diff --git a/.github/workflows/ci.yml b/.github/workflows/ci.yml index f327ddfb8..7ebc28f5b 100644 --- a/.github/workflows/ci.yml +++ b/.github/workflows/ci.yml @@ -160,17 +160,14 @@ jobs: run: node scripts/check-rust-kernel-parity.mjs --api-only authoring-smoke: - # Drift guard: scaffold -> install -> build -> harness end to end. Reds the - # build the moment the `runx new` authoring path breaks (stale package pins, - # toolkit drift, broken templates), so it can never silently rot again. + # Drift guard: scaffold -> harness end to end. Reds the build the moment the + # `runx new` native scaffold breaks (broken templates, harness regressions), + # so the authoring path can never silently rot. Native skills carry no npm + # deps and no build step, so this is scaffold then harness, nothing else. runs-on: ubuntu-latest steps: - name: Checkout uses: actions/checkout@v6 - - name: Setup pnpm - uses: pnpm/action-setup@v5 - with: - version: 10.18.2 - name: Setup Node uses: actions/setup-node@v6 with: @@ -184,7 +181,7 @@ jobs: - name: Build runx working-directory: crates run: cargo build -p runx-cli - - name: Scaffold, install, build, harness + - name: Scaffold and harness env: RUNX_RECEIPT_SIGN_KID: ci-authoring-smoke RUNX_RECEIPT_SIGN_ED25519_SEED_BASE64: QkJCQkJCQkJCQkJCQkJCQkJCQkJCQkJCQkJCQkJCQkI= @@ -193,10 +190,7 @@ jobs: set -euo pipefail RUNX="$GITHUB_WORKSPACE/crates/target/debug/runx" "$RUNX" new authoring-smoke --directory "$RUNNER_TEMP/authoring-smoke" - cd "$RUNNER_TEMP/authoring-smoke" - pnpm install - "$RUNX" tool build --all --json - "$RUNX" harness . --json + "$RUNX" harness "$RUNNER_TEMP/authoring-smoke" --json checks: runs-on: ubuntu-latest diff --git a/.github/workflows/code-review-note-harness.yml b/.github/workflows/code-review-note-harness.yml new file mode 100644 index 000000000..c0b200c42 --- /dev/null +++ b/.github/workflows/code-review-note-harness.yml @@ -0,0 +1,36 @@ +name: code-review-note harness + +on: + workflow_dispatch: + push: + branches: + - code-review-note + paths: + - 'skills/code-review-note/**' + - '.github/workflows/code-review-note-harness.yml' + +jobs: + harness: + runs-on: ubuntu-latest + steps: + - uses: actions/checkout@v4 + - uses: actions/setup-node@v4 + with: + node-version: '22' + - name: Run local harness on Linux + env: + RUNX_RECEIPT_SIGN_KID: runx-demo-key + RUNX_RECEIPT_SIGN_ED25519_SEED_BASE64: QkJCQkJCQkJCQkJCQkJCQkJCQkJCQkJCQkJCQkJCQkI= + RUNX_RECEIPT_SIGN_ISSUER_TYPE: hosted + run: | + mkdir -p /tmp/runx-code-review-note-receipts + npx --yes @runxhq/cli@0.6.13 --version + npx --yes @runxhq/cli@0.6.13 harness ./skills/code-review-note --receipt-dir /tmp/runx-code-review-note-receipts --json | tee code-review-note-harness.json + test "$(node -e "const r=require('./code-review-note-harness.json'); process.stdout.write(r.status)")" = passed + - name: Upload harness evidence + uses: actions/upload-artifact@v4 + with: + name: code-review-note-harness-evidence + path: | + code-review-note-harness.json + /tmp/runx-code-review-note-receipts/*.json diff --git a/.github/workflows/frantic-github-thread-sync.yml b/.github/workflows/frantic-github-thread-sync.yml index c83005750..d2c088ade 100644 --- a/.github/workflows/frantic-github-thread-sync.yml +++ b/.github/workflows/frantic-github-thread-sync.yml @@ -4,6 +4,12 @@ env: FORCE_JAVASCRIPT_ACTIONS_TO_NODE24: true on: + # The venue fires repository_dispatch (board-sync) on board mutations so the drain + # runs promptly. The cron below is the backstop: GitHub throttles high-frequency + # schedules (they fire roughly hourly under load), so this event is what makes the + # sync low-latency and reliable rather than schedule-only. + repository_dispatch: + types: [board-sync] workflow_dispatch: inputs: after_event_id: diff --git a/.github/workflows/release.yml b/.github/workflows/release.yml index 5bab767bd..333f4bf03 100644 --- a/.github/workflows/release.yml +++ b/.github/workflows/release.yml @@ -307,25 +307,6 @@ jobs: fi ( cd "$dir" && npm publish --access public --tag latest ) done - - name: Publish workspace JS packages - # The cli-v* release also publishes the workspace JS packages the `runx new` - # authoring toolkit depends on, so they can never lag the binary again - # (the prior drift root cause). pnpm publish resolves the workspace: - # protocol to concrete versions; the list is dep-ordered (contracts first). - env: - NODE_AUTH_TOKEN: ${{ secrets.NPM_TOKEN }} - NPM_CONFIG_PROVENANCE: "true" - run: | - set -euo pipefail - for dir in packages/contracts packages/authoring; do - name=$(node -p "require('./${dir}/package.json').name") - version=$(node -p "require('./${dir}/package.json').version") - if npm view "${name}@${version}" version >/dev/null 2>&1; then - echo "${name}@${version} already published; skipping" - continue - fi - pnpm --filter "${name}" publish --no-git-checks --access public --tag latest - done publish-crates: needs: [prepare, github-release] diff --git a/CONTRIBUTING.md b/CONTRIBUTING.md index 9b0bc4636..144e107e7 100644 --- a/CONTRIBUTING.md +++ b/CONTRIBUTING.md @@ -120,13 +120,7 @@ Use `runx new ` when you already have the runx CLI available locally and w runx new docs-demo ``` -Use `npm create @runxhq/skill@latest ` for a cold start from npm: - -```bash -npm create @runxhq/skill@latest docs-demo -``` - -Both entry points go through the same scaffolder. Community skills should be authored as standalone packages; the runx repo itself is the first-party lane for official skills, runtime code, tests, and examples. +Community skills should be authored as standalone packages; the runx repo itself is the first-party lane for official skills, runtime code, tests, and examples. The first runnable example is documented in [docs/getting-started.md](docs/getting-started.md). The generated package export index is in [docs/api-surface.md](docs/api-surface.md). diff --git a/README.md b/README.md index 52810ceac..60673853d 100644 --- a/README.md +++ b/README.md @@ -1,8 +1,8 @@

runx

-

the governed runtime for agent skills

+

force multiplier for AI agents

-

expertise as a URL, run under the authority you grant, sealed in a receipt you can replay.

+

Composable skill chains, governed authority, verifiable receipts.

license: MIT @@ -14,9 +14,15 @@ --- -Agents are getting capable faster than we can trust them. They write code, move money, and reach into production. The missing piece is not more intelligence. It is a way to hand an agent a capability and still answer for what it did with it. +runx turns expertise into portable agent infrastructure. A skill is a +`SKILL.md` published at a URL; agents can pull it into their own environment, +compose it with other skills, and build chains of useful work without bespoke +glue code. -runx is that layer. A skill is a `SKILL.md` you publish at a URL. Drop the URL into any agent and it runs in your environment, under the authority you grant, and every step seals into a signed receipt you can replay months later. +That power needs a boundary. runx admits each act under explicit authority, +delivers credentials without turning them into prompt material, runs the +declared profile, and seals the result into a receipt. Authority narrows through +the chain, so agent work can compound without becoming ambient trust. ```text a skill is a URL. @@ -27,29 +33,81 @@ every act produces a receipt. ## quickstart +Install the CLI: + ```bash -npm i -g @runxhq/cli # ships the native runx binary +npm i -g @runxhq/cli +# or: curl -fsSL https://runx.ai/install | sh +``` + +Then choose how you want to run skills. + +### agent path + +Paste [runx.ai/SKILL.md](https://runx.ai/SKILL.md) into your agent. It teaches +the agent how to use the runx CLI, discover skills from the catalog at +[runx.ai/x](https://runx.ai/x), and return receipts. + +```text +Use runx to plan and execute end-to-end business ops for my company. +Goal: prepare API v2 for launch. +Stop before sends, spend, merges, deploys, or publishing. Return receipts. ``` -Run the checked-in example and read its receipt. The signing key below is a public demo key, for local smoke tests only: +### CLI path + +Run any catalog skill directly: ```bash -git clone https://github.com/runxhq/runx && cd runx/oss +runx skill -i key=value --json +``` -export RUNX_RECEIPT_SIGN_KID=runx-demo-key -export RUNX_RECEIPT_SIGN_ED25519_SEED_BASE64=QkJCQkJCQkJCQkJCQkJCQkJCQkJCQkJCQkJCQkJCQkI= -export RUNX_RECEIPT_SIGN_ISSUER_TYPE=hosted -export RUNX_RECEIPT_DIR="$(mktemp -d)" +`business-ops` is one prebuilt skill for managing a business goal: -runx skill examples/hello-world --message "hello from runx" --json # -> status: "sealed" -runx history # what ran, under what authority, with what result +```bash +runx skill business-ops \ + -i signal="prepare API v2 for launch: docs, release, customer comms, issue-to-PR, spend review, and audit" \ + --json ``` -Full walkthrough, including production signing keys, is in [docs/getting-started.md](docs/getting-started.md). +![Basic runx business ops graph](docs/assets/ops-fanout.svg) -## a skill is a URL +The graph is the core shape: goal in, governed lanes out, receipts and approval +gates back. Real teams replace the demo lanes with private context, policies, +tools, and readbacks. -A skill is one file: prose for the model, a typed execution profile for the runtime. +Some other examples: + +```bash +# Docs/product engineering: plan, author, build, critique, and verify docs. +runx skill sourcey -i project=. --json + +# Research/strategy: produce a governed decision brief. +runx skill deep-research-brief \ + -i objective="Which launch risks should we resolve first?" \ + --json + +# Maintainer ops: draft a useful issue response. +runx skill issue-triage \ + -i issue_url=https://github.com/runxhq/runx/issues/241 \ + -i objective="Draft the next helpful maintainer response" \ + --json +``` + +For payment or spend lanes, inspect the skill before granting authority: + +```bash +runx registry search payments +runx registry read runx/x402-pay@sha-008aef3f3b2e +``` + +## skills and execution profiles + +A skill is expertise as a URL. It starts as a portable `SKILL.md`: prose for +the model and a human-readable contract for the operator. When the skill needs +deterministic runners, typed inputs, graph stages, receipt mapping, harness +cases, or governed side effects, it also carries an execution profile named +`X.yaml`. ```yaml --- @@ -60,66 +118,220 @@ source: command: node args: [run.mjs] sandbox: - profile: readonly # what it is allowed to touch + profile: readonly cwd_policy: skill-directory inputs: message: { type: string, required: true } +runx: + category: ops --- Print one message so a new contributor can verify the local runx execution path. ``` -The prose tells the agent what to do. The frontmatter tells runx what it is allowed to do. Publish it and the URL is the skill. Browse the open catalog at [runx.ai/x](https://runx.ai/x). +`SKILL.md` is the capability. `X.yaml` is the execution profile. Keep the +profile explicit: runner wiring, typed inputs and outputs, tool/context refs, +authority and receipt mapping, side-effect posture, and harness cases. Do not +use it as a strategy document, private state file, target registry, copy deck, +or secret container. + +Browse the public catalog at [runx.ai/x](https://runx.ai/x). + +## graphs make acts composable + +Graphs let one governed act consume the receipt-backed output of another: + +```yaml +name: hello-graph +owner: runx +steps: + - id: first + skill: ../hello-world + inputs: + message: hello from graph + - id: second + skill: ../hello-world + context: + message: first.stdout +``` + +The important boundary is not "how many model calls happened." The boundary is +what must be guaranteed. Agents are for judgment and authoring. Required +mutations, API calls, payments, and provider writes belong in deterministic +steps, where the runtime can admit the authority, perform the act, and seal the +result. + +Use a graph when phases, approvals, or side effects need to be visible in the +execution record. Use one bounded skill run when a single act is enough. + +See [docs/skill-to-graph.md](docs/skill-to-graph.md). + +## authority without secret leakage -## the model +runx receipts explain the authority boundary without becoming a secret side +channel. -Nine objects, one runtime. A run is a graph; every hop runs the same four steps, and authority only narrows as it descends. +Public proof may include: -- **skill**: expertise plus a typed execution profile. -- **graph**: skills calling skills. runx renders the topology from the skills themselves and scopes authority at every branch, with no orchestration layer to maintain. -- **bounds**: least privilege by default. Grants are explicit, and an over-scope request is refused before anything runs. -- **receipt**: every act is signed and linked into one reproducible record. The artifact a CISO accepts and a developer can replay. +- requested scopes, granted scopes, grant id, and admission decision; +- provider, connection id, grant reference, and credential material hash; +- sandbox profile, declared enforcement, runtime enforcer, and approval result; +- redaction status and output hashes. -The full grammar (the four-step hop, guards, conditional `when` branches, the act model) lives in the [spec](https://runx.ai/spec). +Public proof must not include: -## three things you couldn't do before +- raw access tokens, refresh tokens, API keys, passwords, or client secrets; +- full private stdout or stderr bodies; +- ambient environment dumps; +- unchecked provider output bodies in public evidence. -- **expertise ships as a link.** A skill is a URL any agent can run in its own environment, under its own grants and approval gates. -- **graphs compose themselves.** One skill calls another, which calls a third. The topology comes from the skills, not from glue code you maintain. -- **receipts are proof.** Signed, linked, replayable. Reputation becomes something you verify instead of something you take on faith. +Provider-permission effects fail closed unless the operator supplies explicit +grant evidence. Spend-class payment authority must carry aggregate caps, not +only per-call limits. `runx doctor authority --json` reports readiness without +printing secret values. -## run it yourself +See [docs/security-authority-proof.md](docs/security-authority-proof.md). -runx is MIT and runs entirely in your environment. Your keys, your boundary; your data and network never leave your control. The trusted local runtime is Rust, with no hosted dependency for local execution: +## demos that prove boundaries + +These demos are runnable from this repo and produce receipts: + +| Demo | What it proves | Run | +| --- | --- | --- | +| `examples/hello-world` | Native CLI skill path, sealed receipt baseline | `runx harness examples/hello-world` | +| `skills/business-ops` | One business signal fans out through governed ops lanes and seals a graph receipt | `runx harness skills/business-ops` | +| `examples/github-mcp-hero` | Governed GitHub read succeeds, out-of-scope write is refused, denial receipt verifies | `sh examples/github-mcp-hero/run.sh` | +| `examples/http-graph` | Governed HTTP front call against a local fixture seals a receipt tree | `sh examples/http-graph/run.sh` | +| `examples/openapi-graph` | OpenAPI operation runs through the external-adapter lane and seals | `sh examples/openapi-graph/run.sh` | +| `examples/governed-spend/skills/overspend-refused` | Spend above authority is refused before rail execution | `runx harness examples/governed-spend/skills/overspend-refused` | +| `examples/loop-orchestration` | Bounded outer loop submits governed turns, prints receipt ids, and demonstrates refusal | `sh examples/loop-orchestration/run.sh` | + +For deterministic payment dogfood without funded wallets or provider keys: ```bash -cd oss && cargo build --manifest-path crates/Cargo.toml -p runx-cli +pnpm demos:check ``` -`@runxhq/cli` is the published distribution of that same binary. +See [docs/demos.md](docs/demos.md). + +## what a receipt proves + +A runx receipt is designed to answer the questions that matter after the agent +has moved on: + +| Question | Receipt surface | +| --- | --- | +| What ran? | `subject`, skill ref, source type, runner metadata | +| Who or what admitted it? | `authority.actor_ref`, grant refs, authority proof refs | +| What was allowed? | requested scopes, granted scopes, sandbox policy, approval metadata | +| What happened? | act entries, output artifacts, exit status, closure summary | +| Can it be checked later? | content-addressed id, canonical digest, signature, lineage | +| Did secrets leak into proof? | redacted metadata, hashed material refs, banned raw credential bodies | + +Shape, abbreviated: + +```json +{ + "schema": "runx.receipt.v1", + "subject": { "kind": "skill" }, + "authority": { + "actor_ref": { "type": "principal", "uri": "runx:principal:local_runtime" }, + "grant_refs": [] + }, + "seal": { + "disposition": "closed", + "reason_code": "process_closed" + }, + "lineage": { + "parent": null, + "children": [] + } +} +``` -## author and publish +Offline verification recomputes the canonical body digest, checks the +content-addressed id, verifies signatures when trusted keys are configured, and +can walk receipt ancestry from a receipt store. + +The receipt is not the product by itself. It is where authority, action, +evidence, and future learning meet in one verifiable object. + +## governed execution invariant + +Every governed execution passes through the same four stages: + +```text +admit -> deliver credentials -> sandbox -> seal +``` + +| Stage | What runx protects | +| --- | --- | +| `admit` | Policy checks the requested act before any step handler runs. An unadmitted act never reaches execution. | +| `deliver credentials` | Secret material crosses only a structured delivery boundary. Receipts carry grant refs, public observations, and hashes, not tokens. | +| `sandbox` | The declared cwd, env, filesystem, network, and enforcement posture are resolved and recorded. Runs can fail closed when OS-level enforcement is required. | +| `seal` | The runtime writes a signed `runx.receipt.v1` record with subject, authority witness, outputs, lineage, and closure. | + +## publish and trust + +Community skills should be standalone packages: `SKILL.md`, optional `X.yaml`, +and only the files runx can consume. Publish locally first: + +```bash +runx registry publish ./skills//SKILL.md +``` + +Then publish to the hosted catalog when you want shared discovery: ```bash -runx new my-skill # scaffold a standalone skill package -npm create @runxhq/skill@latest my-skill # or start from npm +runx login --for publish +runx registry publish ./skills//SKILL.md --registry https://api.runx.ai ``` -Write the prose, declare the profile, run it locally, then publish from a public repo at [runx.ai/x/publish](https://runx.ai/x/publish) or with `runx login && runx registry publish`. This repo is the first-party lane for official skills and the runtime; community skills ship as standalone packages. +Hosted publishing reconstructs the submitted package, reruns the harness, rejects +failed cases, and stores immutable package digests. New rows start as +`community`; verification and evidence promote discovery. Publisher declaration +alone is not trust. + +See [docs/publishing.md](docs/publishing.md). + +## architecture + +Rust owns the trusted local runtime path. + +| Layer | Owner | +| --- | --- | +| policy, state machine, authority admission | `runx-core` | +| skill, graph, runner, and tool manifest parsing | `runx-parser` | +| canonical receipts, hashing, signatures, tree verification | `runx-receipts` | +| local runtime, adapters, sandbox planning, harness, registry, MCP, payment gates | `runx-runtime` | +| native binary | `runx-cli` | +| generated TypeScript contracts and npm wrapper | `@runxhq/contracts`, `@runxhq/cli` | + +TypeScript remains for generated contracts, client wrappers, cloud/product +integrations, host adapters, authoring tooling, and helper SDKs. It must not be +a fallback executor for trusted local behavior. + +See [docs/reference.md](docs/reference.md) and +[docs/rust-kernel-architecture.md](docs/rust-kernel-architecture.md). ## docs -| | | +| Read this | When you need | | --- | --- | -| [getting started](docs/getting-started.md) | install, first skill, first receipt | -| [skill to graph](docs/skill-to-graph.md) | compose skills into a governed graph | -| [the spec](https://runx.ai/spec) | the act model, the four-step hop, the grammar | -| [the catalog](https://runx.ai/x) | every governed skill, by URL | -| [architecture and reference](docs/reference.md) | crate topology, sandbox posture, tool authoring, the full surface | +| [getting started](docs/getting-started.md) | first skill, first receipt | +| [skill to graph](docs/skill-to-graph.md) | compose governed acts | +| [security authority proof](docs/security-authority-proof.md) | scope, credentials, grants, verification | +| [demos](docs/demos.md) | runnable proof paths | +| [publishing](docs/publishing.md) | local and hosted skill publishing | +| [reference](docs/reference.md) | CLI, crates, registry, receipts, extension protocols | +| [the spec](https://runx.ai/spec) | act model, receipt grammar, public contracts | +| [the catalog](https://runx.ai/x) | governed skills by URL | ## contributing -Setup, test selection, and sign-off rules are in [CONTRIBUTING.md](CONTRIBUTING.md). Security policy: [SECURITY.md](SECURITY.md). runx is MIT licensed; see [LICENSE](LICENSE). +Setup, test selection, and sign-off rules are in +[CONTRIBUTING.md](CONTRIBUTING.md). Security policy: +[SECURITY.md](SECURITY.md). runx is MIT licensed; see [LICENSE](LICENSE). --- diff --git a/crates/Cargo.lock b/crates/Cargo.lock index 355252635..aa0b0801e 100644 --- a/crates/Cargo.lock +++ b/crates/Cargo.lock @@ -2,6 +2,12 @@ # It is not intended for manual editing. version = 4 +[[package]] +name = "adler2" +version = "2.0.1" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "320119579fcad9c21884f5c4861d16174d0e06250625266f50fe6898340abefa" + [[package]] name = "aead" version = "0.5.2" @@ -60,6 +66,21 @@ dependencies = [ "memchr", ] +[[package]] +name = "alloc-no-stdlib" +version = "2.0.4" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "cc7bb162ec39d46ab1ca8c77bf72e890535becd1751bb45f64c597edb4c8c6b3" + +[[package]] +name = "alloc-stdlib" +version = "0.2.4" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "0e76a019e91224d279006ff972f1e984179a6e9feb050adba6ce8274aef23195" +dependencies = [ + "alloc-no-stdlib", +] + [[package]] name = "allocator-api2" version = "0.2.21" @@ -87,6 +108,18 @@ version = "1.0.14" source = "registry+https://github.com/rust-lang/crates.io-index" checksum = "940b3a0ca603d1eade50a4846a2afffd5ef57a9feac2c0e2ec2e14f9ead76000" +[[package]] +name = "async-compression" +version = "0.4.42" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "e79b3f8a79cccc2898f31920fc69f304859b3bd567490f75ebf51ae1c792a9ac" +dependencies = [ + "compression-codecs", + "compression-core", + "pin-project-lite", + "tokio", +] + [[package]] name = "async-trait" version = "0.1.89" @@ -152,6 +185,27 @@ version = "0.2.4" source = "registry+https://github.com/rust-lang/crates.io-index" checksum = "dc0b364ead1874514c8c2855ab558056ebfeb775653e7ae45ff72f28f8f3166c" +[[package]] +name = "brotli" +version = "8.0.4" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "5cc91aac060a7a1e25823bdccbfb6af1875b88f17c6daac97894eed8207166b3" +dependencies = [ + "alloc-no-stdlib", + "alloc-stdlib", + "brotli-decompressor", +] + +[[package]] +name = "brotli-decompressor" +version = "5.0.3" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "3a32acac15fe1967bc3986b2a6347dffc965602354ea6f450ad07e8bfd253583" +dependencies = [ + "alloc-no-stdlib", + "alloc-stdlib", +] + [[package]] name = "bumpalo" version = "3.20.3" @@ -183,6 +237,8 @@ source = "registry+https://github.com/rust-lang/crates.io-index" checksum = "dad887fd958be91b5098c0248def011f4523ab786cd411be668777e55063501f" dependencies = [ "find-msvc-tools", + "jobserver", + "libc", "shlex", ] @@ -287,6 +343,26 @@ dependencies = [ "memchr", ] +[[package]] +name = "compression-codecs" +version = "0.4.38" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "ce2548391e9c1929c21bf6aa2680af86fe4c1b33e6cea9ac1cfeec0bd11218cf" +dependencies = [ + "brotli", + "compression-core", + "flate2", + "memchr", + "zstd", + "zstd-safe", +] + +[[package]] +name = "compression-core" +version = "0.4.32" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "cc14f565cf027a105f7a44ccf9e5b424348421a1d8952a8fc9d499d313107789" + [[package]] name = "core-foundation" version = "0.10.1" @@ -321,6 +397,15 @@ dependencies = [ "libc", ] +[[package]] +name = "crc32fast" +version = "1.5.0" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "9481c1c90cbf2ac953f07c8d4a58aa3945c425b7185c9154d67a65e4230da511" +dependencies = [ + "cfg-if", +] + [[package]] name = "criterion" version = "0.5.1" @@ -495,6 +580,16 @@ version = "0.1.9" source = "registry+https://github.com/rust-lang/crates.io-index" checksum = "5baebc0774151f905a1a2cc41989300b1e6fbb29aff0ceffa1064fdd3088d582" +[[package]] +name = "flate2" +version = "1.1.9" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "843fba2746e448b37e26a819579957415c8cef339bf08564fe8b7ddbd959573c" +dependencies = [ + "crc32fast", + "miniz_oxide", +] + [[package]] name = "fluent-uri" version = "0.4.1" @@ -506,6 +601,12 @@ dependencies = [ "serde", ] +[[package]] +name = "fnv" +version = "1.0.7" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "3f9eec918d3f24069decb9af1554cad7c880e2da24a9afd88aca000531ab82c1" + [[package]] name = "foldhash" version = "0.2.0" @@ -676,6 +777,25 @@ dependencies = [ "polyval", ] +[[package]] +name = "h2" +version = "0.4.14" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "171fefbc92fe4a4de27e0698d6a5b392d6a0e333506bc49133760b3bcf948733" +dependencies = [ + "atomic-waker", + "bytes", + "fnv", + "futures-core", + "futures-sink", + "http", + "indexmap", + "slab", + "tokio", + "tokio-util", + "tracing", +] + [[package]] name = "half" version = "2.7.1" @@ -765,6 +885,7 @@ dependencies = [ "bytes", "futures-channel", "futures-core", + "h2", "http", "http-body", "httparse", @@ -1040,6 +1161,16 @@ dependencies = [ "syn", ] +[[package]] +name = "jobserver" +version = "0.1.34" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "9afb3de4395d6b3e67a780b6de64b51c978ecf11cb9a462c66be7d4ca9039d33" +dependencies = [ + "getrandom 0.3.4", + "libc", +] + [[package]] name = "js-sys" version = "0.3.102" @@ -1129,6 +1260,16 @@ version = "0.3.0" source = "registry+https://github.com/rust-lang/crates.io-index" checksum = "c2a86d3146ed3995b5913c414f6664344b9617457320782e64f0bb44afd49d74" +[[package]] +name = "miniz_oxide" +version = "0.8.9" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "1fa76a2c86f704bdb222d66965fb3d63269ce38518b83cb0575fca855ebb6316" +dependencies = [ + "adler2", + "simd-adler32", +] + [[package]] name = "mio" version = "1.2.1" @@ -1290,6 +1431,12 @@ version = "0.2.17" source = "registry+https://github.com/rust-lang/crates.io-index" checksum = "a89322df9ebe1c1578d689c92318e070967d1042b512afbe49518723f4e6d5cd" +[[package]] +name = "pkg-config" +version = "0.3.33" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "19f132c84eca552bf34cab8ec81f1c1dcc229b811638f9d283dceabe58c5569e" + [[package]] name = "plotters" version = "0.3.7" @@ -1561,6 +1708,7 @@ dependencies = [ "base64", "bytes", "futures-core", + "h2", "http", "http-body", "http-body-util", @@ -1633,7 +1781,7 @@ dependencies = [ [[package]] name = "runx-cli" -version = "0.6.0" +version = "0.6.6" dependencies = [ "base64", "ring", @@ -2048,6 +2196,12 @@ dependencies = [ "libc", ] +[[package]] +name = "simd-adler32" +version = "0.3.9" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "703d5c7ef118737c72f1af64ad2f6f8c5e1921f818cdcb97b8fe6fc69bf66214" + [[package]] name = "simd_cesu8" version = "1.1.1" @@ -2277,12 +2431,17 @@ version = "0.6.11" source = "registry+https://github.com/rust-lang/crates.io-index" checksum = "4cfcf7e2740e6fc6d4d688b4ef00650406bb94adf4731e43c096c3a19fe40840" dependencies = [ + "async-compression", "bitflags", "bytes", + "futures-core", "futures-util", "http", "http-body", + "http-body-util", "pin-project-lite", + "tokio", + "tokio-util", "tower", "tower-layer", "tower-service", @@ -2822,3 +2981,31 @@ name = "zmij" version = "1.0.21" source = "registry+https://github.com/rust-lang/crates.io-index" checksum = "b8848ee67ecc8aedbaf3e4122217aff892639231befc6a1b58d29fff4c2cabaa" + +[[package]] +name = "zstd" +version = "0.13.3" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "e91ee311a569c327171651566e07972200e76fcfe2242a4fa446149a3881c08a" +dependencies = [ + "zstd-safe", +] + +[[package]] +name = "zstd-safe" +version = "7.2.4" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "8f49c4d5f0abb602a93fb8736af2a4f4dd9512e36f7f570d66e65ff867ed3b9d" +dependencies = [ + "zstd-sys", +] + +[[package]] +name = "zstd-sys" +version = "2.0.16+zstd.1.5.7" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "91e19ebc2adc8f83e43039e79776e3fda8ca919132d68a1fed6a5faca2683748" +dependencies = [ + "cc", + "pkg-config", +] diff --git a/crates/deny.toml b/crates/deny.toml index 879f0b9b2..92b5608f6 100644 --- a/crates/deny.toml +++ b/crates/deny.toml @@ -30,7 +30,7 @@ deny = [ # keeps rmcp out of pure crates and the CLI. { name = "serde_yaml", reason = "Unapproved YAML backend; the parser backend is serde_norway." }, { name = "serde_yml", reason = "Retired YAML backend candidate; the parser backend is serde_norway." }, - { name = "tokio", wrappers = ["runx-runtime", "reqwest", "rmcp", "hyper", "hyper-rustls", "hyper-util", "tokio-rustls", "tokio-stream", "tokio-util", "tower"], reason = "Approved only inside runx-runtime async-http/MCP adapter boundaries and reviewed transport internals; pure crates must not depend on tokio." }, + { name = "tokio", wrappers = ["runx-runtime", "reqwest", "rmcp", "async-compression", "h2", "hyper", "hyper-rustls", "hyper-util", "tokio-rustls", "tokio-stream", "tokio-util", "tower", "tower-http"], reason = "Approved only inside runx-runtime async-http/MCP adapter boundaries and reviewed transport internals; pure crates must not depend on tokio." }, { name = "ureq", reason = "No HTTP client exception is approved; adapter-side HTTP needs a scoped spec first." }, ] diff --git a/crates/runx-cli/Cargo.toml b/crates/runx-cli/Cargo.toml index 5b284875e..fa3a39a5d 100644 --- a/crates/runx-cli/Cargo.toml +++ b/crates/runx-cli/Cargo.toml @@ -5,7 +5,7 @@ autotests = false name = "runx-cli" # Kept in lockstep with packages/cli/package.json (the npm distribution line). # The release workflow stamps this from the npm manifest before building. -version = "0.6.0" +version = "0.6.6" edition.workspace = true rust-version.workspace = true description = "Cargo-installed launcher for the runx governed agent workflow CLI." diff --git a/crates/runx-cli/src/config.rs b/crates/runx-cli/src/config.rs index 7d0325ff2..3eb754522 100644 --- a/crates/runx-cli/src/config.rs +++ b/crates/runx-cli/src/config.rs @@ -102,7 +102,7 @@ pub fn parse_config_plan(args: &[OsString]) -> Result { let mut index = 2; while index < args.len() { let token = os_arg(args, index, "config")?; - if !token.starts_with("--") { + if !token.starts_with('-') { positionals.push(token.to_owned()); index += 1; continue; @@ -110,7 +110,7 @@ pub fn parse_config_plan(args: &[OsString]) -> Result { let (flag, inline_value) = split_flag(token); match flag { - "--json" => { + "--json" | "-j" => { if inline_value.is_some() { return Err("--json does not take a value".to_owned()); } @@ -139,7 +139,7 @@ pub fn parse_config_plan(args: &[OsString]) -> Result { }; Ok(ConfigPlan { action, - key: Some(key.clone()), + key: Some(normalize_config_key(key).to_owned()), value: None, json, }) @@ -153,7 +153,7 @@ pub fn parse_config_plan(args: &[OsString]) -> Result { } Ok(ConfigPlan { action, - key: Some(key.clone()), + key: Some(normalize_config_key(key).to_owned()), value: Some(values.join(" ")), json, }) @@ -161,6 +161,16 @@ pub fn parse_config_plan(args: &[OsString]) -> Result { } } +fn normalize_config_key(key: &str) -> &str { + match key { + "provider" => "agent.provider", + "model" => "agent.model", + "api-key" | "agent-key" => "agent.api_key", + "public-token" => "public.api_token", + _ => key, + } +} + pub fn run_config_command( plan: &ConfigPlan, env: &BTreeMap, @@ -305,10 +315,10 @@ mod tests { parse_config_plan(&[ "config".into(), "set".into(), - "agent.model".into(), + "model".into(), "gpt".into(), "test".into(), - "--json".into(), + "-j".into(), ]), Ok(ConfigPlan { action: ConfigAction::Set, diff --git a/crates/runx-cli/src/doctor.rs b/crates/runx-cli/src/doctor.rs index 53b66af42..1a55f492a 100644 --- a/crates/runx-cli/src/doctor.rs +++ b/crates/runx-cli/src/doctor.rs @@ -16,7 +16,8 @@ use runx_pay::state::{ use runx_runtime::{ PROVIDER_PERMISSION_GRANT_ID_ENV, PROVIDER_PERMISSION_GRANTED_SCOPES_ENV, RUNX_RECEIPT_SIGN_ED25519_SEED_BASE64_ENV, RUNX_RECEIPT_SIGN_ISSUER_TYPE_ENV, - RUNX_RECEIPT_SIGN_KID_ENV, RuntimeError, default_doctor_options, run_doctor, + RUNX_RECEIPT_SIGN_KID_ENV, RuntimeError, default_doctor_options, load_runx_config_file, + resolve_runx_home_dir, run_doctor, }; use crate::history::{ @@ -76,7 +77,18 @@ fn run_doctor_command( } let root = resolve_doctor_root(plan, env, cwd); - let report = run_doctor(&root, &default_doctor_options())?; + let mut report = run_doctor(&root, &default_doctor_options())?; + if let Some(diagnostic) = managed_agent_config_diagnostic(env, cwd) { + report.diagnostics.push(diagnostic); + } + report.summary = summary(&report.diagnostics); + if report + .diagnostics + .iter() + .any(|diagnostic| diagnostic.severity == DoctorDiagnosticSeverity::Error) + { + report.status = DoctorStatus::Failure; + } let exit_code = match report.status { DoctorStatus::Success => 0, DoctorStatus::Failure => 1, @@ -89,6 +101,160 @@ fn run_doctor_command( Ok(DoctorCliOutput { stdout, exit_code }) } +// rust-style-allow: long-function - this builds one structured diagnostic packet +// from env, config, and credential state so the evidence and repair stay together. +fn managed_agent_config_diagnostic( + env: &BTreeMap, + cwd: &Path, +) -> Option { + let config_dir = resolve_runx_home_dir(env, cwd); + let config_path = config_dir.join("config.json"); + let config = load_runx_config_file(&config_path); + let mut evidence = JsonObject::new(); + evidence.insert( + "config_path".to_owned(), + JsonValue::String(config_path.display().to_string()), + ); + + let (config_provider, config_model, config_key_ref, config_error) = match config { + Ok(config) => ( + config + .agent + .as_ref() + .and_then(|agent| agent.provider.as_deref()) + .map(str::to_owned), + config + .agent + .as_ref() + .and_then(|agent| agent.model.as_deref()) + .map(str::to_owned), + config + .agent + .as_ref() + .and_then(|agent| agent.api_key_ref.as_deref()) + .map(str::to_owned), + None, + ), + Err(error) => (None, None, None, Some(error.to_string())), + }; + + let provider = first_non_empty([ + env.get("RUNX_AGENT_PROVIDER").map(String::as_str), + config_provider.as_deref(), + ]); + let model = first_non_empty([ + env.get("RUNX_AGENT_MODEL").map(String::as_str), + config_model.as_deref(), + ]); + let provider_key_env = provider.and_then(provider_api_key_env); + let api_key_configured = env_contains_non_empty(env, "RUNX_AGENT_API_KEY") + || provider_key_env.is_some_and(|name| env_contains_non_empty(env, name)) + || config_key_ref + .as_deref() + .is_some_and(|value| !value.trim().is_empty()); + + evidence.insert( + "provider_set".to_owned(), + JsonValue::Bool(provider.is_some()), + ); + evidence.insert("model_set".to_owned(), JsonValue::Bool(model.is_some())); + evidence.insert( + "api_key_set".to_owned(), + JsonValue::Bool(api_key_configured), + ); + if let Some(provider) = provider { + evidence.insert( + "provider".to_owned(), + JsonValue::String(provider.to_owned()), + ); + } + if let Some(model) = model { + evidence.insert("model".to_owned(), JsonValue::String(model.to_owned())); + } + if let Some(name) = provider_key_env { + evidence.insert( + "provider_api_key_env".to_owned(), + JsonValue::String(name.to_owned()), + ); + } + if let Some(error) = config_error.as_ref() { + evidence.insert("config_error".to_owned(), JsonValue::String(error.clone())); + } + + let complete = provider.is_some() && model.is_some() && api_key_configured; + let partial = !complete + && (provider.is_some() || model.is_some() || api_key_configured || config_error.is_some()); + if !complete && !partial { + return None; + } + let severity = if partial { + DoctorDiagnosticSeverity::Warning + } else { + DoctorDiagnosticSeverity::Info + }; + let message = if let Some(error) = config_error { + format!("Managed-agent config could not be read: {error}.") + } else if complete { + "Managed-agent config is complete; agent-task runners can execute in-process.".to_owned() + } else { + "Managed-agent config is partial; set provider, model, and API key or unset the partial values. Otherwise agent-task runners may yield to the host or fail later.".to_owned() + }; + + Some(DoctorDiagnostic { + id: "runx.agent.config".to_owned(), + instance_id: "runx:doctor:runx.agent.config".to_owned(), + severity, + title: "Managed-agent config".to_owned(), + message, + target: object([ + ("kind", string_value("config")), + ("ref", string_value("runx.agent.config")), + ]), + location: DoctorLocation { + path: "runx config".to_owned(), + json_pointer: Some("/agent".to_owned()), + }, + evidence: Some(evidence), + repairs: if partial { + vec![DoctorRepair { + id: "runx.agent.config.configure".to_owned(), + kind: DoctorRepairKind::Manual, + confidence: DoctorRepairConfidence::High, + risk: DoctorRepairRisk::Low, + path: Some("runx config".to_owned()), + json_pointer: Some("/agent".to_owned()), + contents: Some( + "Set agent.provider, agent.model, and agent.api_key, or unset partial managed-agent config." + .to_owned(), + ), + patch: None, + command: Some( + "runx config set agent.provider anthropic && runx config set agent.model && runx config set agent.api_key ".to_owned(), + ), + requires_human_review: false, + }] + } else { + Vec::new() + }, + }) +} + +fn first_non_empty<'a>(values: impl IntoIterator>) -> Option<&'a str> { + values + .into_iter() + .flatten() + .map(str::trim) + .find(|value| !value.is_empty()) +} + +fn provider_api_key_env(provider: &str) -> Option<&'static str> { + match provider.trim().to_ascii_lowercase().as_str() { + "anthropic" => Some("ANTHROPIC_API_KEY"), + "openai" => Some("OPENAI_API_KEY"), + _ => None, + } +} + fn run_registry_doctor(env: &BTreeMap, cwd: &Path) -> DoctorReport { let target = registry::resolve_registry_target(®istry_probe_plan(), env, cwd); let diagnostics = vec![ diff --git a/crates/runx-cli/src/export/shim.rs b/crates/runx-cli/src/export/shim.rs index 04dcc870f..9d4550cce 100644 --- a/crates/runx-cli/src/export/shim.rs +++ b/crates/runx-cli/src/export/shim.rs @@ -132,7 +132,12 @@ Interpret the runx JSON result exactly: {{ \"answers\": {{ \"\": {{ - \"...\": \"object matching request.invocation.envelope.output\" + \"...\": \"object matching request.invocation.envelope.output\", + \"closure\": {{ + \"disposition\": \"closed\", + \"reason_code\": \"completed\", + \"summary\": \"concise outcome summary\" + }} }} }} }} diff --git a/crates/runx-cli/src/launcher.rs b/crates/runx-cli/src/launcher.rs index 5fb270fd7..53fdaa918 100644 --- a/crates/runx-cli/src/launcher.rs +++ b/crates/runx-cli/src/launcher.rs @@ -322,13 +322,13 @@ Usage: Commands: runx new [--directory dir] [--json] runx init [-g|--global] [--prefetch official] [--json] - runx verify [receipt-id] [--receipt-dir dir] [--receipt ] [--notary --notary-key trusted.pem] [--json] + runx verify [receipt-id] [--receipt-dir dir] [--receipt ] [--notary --notary-key trusted.pem] [-j|--json] runx history [query] [--skill s] [--status s] [--source s] [--actor a] [--artifact-type t] [--since iso] [--until iso] [--receipt-dir dir] [--json] - runx list [tools|skills|graphs|packets|overlays] [--ok-only|--invalid-only] [--json] - runx login [--provider github|google|gitlab] [--api-base-url url] [--allow-local-api] [--json] - runx config set|get|list [agent.provider|agent.model|agent.api_key|public.api_token] [value] [--json] + runx list [tools|skills|graphs|packets|overlays] [--ok-only|--invalid-only] [-j|--json] + runx login [--provider github|google|gitlab] [--for default|publish] [--api-url url] [--local-api] [-j|--json] + runx config set|get|list [provider|model|api-key|public-token] [value] [-j|--json] runx policy inspect|lint [--json] - runx publish [--api-base-url url] [--token token] [--allow-local-api] [--json] + runx publish [--api-url url] [--token token] [--local-api] [-j|--json] runx kernel eval --input --json runx payment admission issue --input --json runx parser eval --input --json @@ -336,9 +336,9 @@ Commands: runx dev [root] [--lane lane] [--json] runx export [skill-ref...] [--project] [--json] runx mcp serve [--receipt-dir dir] [--http-listen [addr]] [--http-allow-non-loopback] - runx skill [--registry url|path] [--digest sha256] [--input key=value] [--runner name] [--flag value] [--receipt-dir dir] [--run-id id --answers file] [--json] + runx skill [-p profile] [-i key=value] [-j] [--runner name] [--registry url|path] [--digest sha256] [--flag value] [--credential descriptor --secret-env NAME] [-R dir] [--run-id id --answers file] runx add [--registry url|path] [--version version] [--ref git-ref] [--digest sha256] [--to dir] [--installation-id id] [--api-base-url url] [--json] - runx harness [--receipt-dir dir] [--json] + runx harness [-R dir] [-j|--json] runx tool build |--all [--json] runx tool search [--source source] [--json] runx tool inspect [--source source] [--json] @@ -373,13 +373,15 @@ pub fn publish_help_text() -> String { runx publish Usage: - runx publish [--api-base-url url] [--token token] [--allow-local-api] [--json] + runx publish [--api-url url] [--token token] [--local-api] [-j|--json] Options: - --api-base-url url Public API base URL (default: RUNX_PUBLIC_API_BASE_URL or https://runx.ai) + --api-url url Public API base URL (default: RUNX_PUBLIC_API_BASE_URL or https://api.runx.ai) + --api-base-url url Alias for --api-url --token token Public API token (default: RUNX_PUBLIC_API_TOKEN or runx login) - --allow-local-api Allow loopback/private public API URLs for local dogfood only - --json Print the raw notary response as JSON + --local-api Allow loopback/private public API URLs for local dogfood only + --allow-local-api Alias for --local-api + -j, --json Print the raw notary response as JSON " .to_owned() } @@ -389,14 +391,14 @@ pub fn verify_help_text() -> String { runx verify Usage: - runx verify [receipt-id] [--receipt-dir dir] [--receipt ] [--notary --notary-key trusted.pem] [--json] + runx verify [receipt-id] [--receipt-dir dir] [--receipt ] [--notary --notary-key trusted.pem] [-j|--json] Options: --receipt-dir dir --receipt --notary --notary-key trusted.pem - --json + -j, --json " .to_owned() } @@ -406,18 +408,22 @@ pub fn skill_help_text() -> String { runx skill Usage: - runx skill [--registry url|path] [--digest sha256] [--input key=value] [--runner name] [--flag value] [--receipt-dir dir] [--run-id id --answers file] [--json] + runx skill [-p profile] [-i key=value] [-j] [--runner name] [--registry url|path] [--digest sha256] [--flag value] [--credential descriptor --secret-env NAME] [-R dir] [--run-id id --answers file] Options: + -p, --profile name Use a local credential profile from .runx/credentials.json + -i, --input key=value Set a structured input; repeat for multiple inputs + -R, --receipts dir Write receipts under dir + --receipt-dir dir Alias for --receipts + -j, --json Print machine-readable output + --runner name Select a named runner from X.yaml --registry url|path --digest sha256 - --runner name - --input key=value --flag value - --receipt-dir dir + --credential descriptor One-shot local credential descriptor + --secret-env NAME Env var holding the one-shot credential secret --run-id id --answers file - --json " .to_owned() } @@ -484,6 +490,8 @@ fn mcp_runner_before_serve(args: &[OsString]) -> bool { }) } +// rust-style-allow: long-function - harness flag parsing stays local to the +// launcher boundary so native dispatch does not grow a second parser surface. fn native_harness_plan(args: &[OsString]) -> LauncherAction { let mut fixture_paths = Vec::new(); let mut receipt_dir = None; @@ -494,7 +502,7 @@ fn native_harness_plan(args: &[OsString]) -> LauncherAction { return LauncherAction::Error("harness arguments must be UTF-8".to_owned()); }; - if !token.starts_with("--") { + if !token.starts_with('-') { fixture_paths.push(args[index].clone()); index += 1; continue; @@ -502,13 +510,13 @@ fn native_harness_plan(args: &[OsString]) -> LauncherAction { let (flag, inline_value) = split_flag(token); match flag { - "--json" => { + "--json" | "-j" => { if inline_value.is_some() { return LauncherAction::Error("--json does not take a value".to_owned()); } index += 1; } - "--receipt-dir" => match inline_value { + "--receipt-dir" | "--receipts" => match inline_value { Some(value) => { receipt_dir = Some(OsString::from(value)); index += 1; @@ -523,6 +531,18 @@ fn native_harness_plan(args: &[OsString]) -> LauncherAction { index += 2; } }, + "-R" => { + if inline_value.is_some() { + return LauncherAction::Error( + "-R requires a separate directory value".to_owned(), + ); + } + let Some(value) = args.get(index + 1) else { + return LauncherAction::Error("-R requires a directory".to_owned()); + }; + receipt_dir = Some(value.clone()); + index += 2; + } _ => return LauncherAction::Error(format!("unknown harness flag {flag}")), } } @@ -850,7 +870,7 @@ fn parse_doctor_plan(args: &[OsString]) -> Result { while index < args.len() { let token = os_arg(args, index, "doctor")?; - if !token.starts_with("--") { + if !token.starts_with('-') { if matches!(token, "authority" | "registry") && path.is_none() && mode == DoctorMode::Workspace @@ -879,7 +899,7 @@ fn parse_doctor_plan(args: &[OsString]) -> Result { let (flag, inline_value) = split_flag(token); match flag { - "--json" => { + "--json" | "-j" => { if inline_value.is_some() { return Err("--json does not take a value".to_owned()); } @@ -910,7 +930,7 @@ fn parse_list_plan(args: &[OsString]) -> Result { while index < args.len() { let token = os_arg(args, index, "list")?; - if !token.starts_with("--") { + if !token.starts_with('-') { if saw_kind { return Err("runx list accepts at most one kind".to_owned()); } @@ -927,7 +947,7 @@ fn parse_list_plan(args: &[OsString]) -> Result { return Err(format!("{flag} does not take a value")); } let requested = match flag { - "--json" => { + "--json" | "-j" => { json = true; index += 1; continue; @@ -962,49 +982,20 @@ fn parse_kernel_plan(args: &[OsString]) -> Result { if subcommand != "eval" { return Err(format!("unknown kernel subcommand {subcommand}")); } - - let mut input = None; - let mut json = false; - let mut index = 2; - - while index < args.len() { - let token = os_arg(args, index, "kernel")?; - if !token.starts_with("--") { - return Err(format!("unexpected kernel eval argument {token}")); - } - - let (flag, inline_value) = split_flag(token); - match flag { - "--json" => { - if inline_value.is_some() { - return Err("--json does not take a value".to_owned()); - } - json = true; - index += 1; - } - "--input" => { - if input.is_some() { - return Err("runx kernel eval accepts exactly one --input".to_owned()); - } - let (value, next_index) = flag_value(args, index, flag, inline_value, "kernel")?; - input = Some(if value == "-" { - KernelInputSource::Stdin - } else { - KernelInputSource::Path(PathBuf::from(value)) - }); - index = next_index; - } - _ => return Err(format!("unknown kernel eval flag {flag}")), - } - } - - if !json { - return Err("runx kernel eval requires --json".to_owned()); - } - + let parsed = parse_json_eval_input( + args, + 2, + JsonEvalCommand { + command: "kernel", + subject: "kernel eval", + duplicate_input: "runx kernel eval accepts exactly one --input", + requires_json: "runx kernel eval requires --json", + requires_input: "runx kernel eval requires --input ", + }, + )?; Ok(KernelPlan { - input: input.ok_or_else(|| "runx kernel eval requires --input ".to_owned())?, - json, + input: parsed.input.into_kernel_source(), + json: true, }) } @@ -1020,57 +1011,21 @@ fn parse_payment_plan(args: &[OsString]) -> Result { if action != "issue" { return Err(format!("unknown payment admission subcommand {action}")); } - - let mut input = None; - let mut json = false; - let mut index = 3; - - while index < args.len() { - let token = os_arg(args, index, "payment admission issue")?; - if !token.starts_with("--") { - return Err(format!( - "unexpected payment admission issue argument {token}" - )); - } - - let (flag, inline_value) = split_flag(token); - match flag { - "--json" => { - if inline_value.is_some() { - return Err("--json does not take a value".to_owned()); - } - json = true; - index += 1; - } - "--input" => { - if input.is_some() { - return Err( - "runx payment admission issue accepts exactly one --input".to_owned() - ); - } - let (value, next_index) = - flag_value(args, index, flag, inline_value, "payment admission issue")?; - input = Some(if value == "-" { - PaymentInputSource::Stdin - } else { - PaymentInputSource::Path(PathBuf::from(value)) - }); - index = next_index; - } - _ => return Err(format!("unknown payment admission issue flag {flag}")), - } - } - - if !json { - return Err("runx payment admission issue requires --json".to_owned()); - } - + let parsed = parse_json_eval_input( + args, + 3, + JsonEvalCommand { + command: "payment admission issue", + subject: "payment admission issue", + duplicate_input: "runx payment admission issue accepts exactly one --input", + requires_json: "runx payment admission issue requires --json", + requires_input: "runx payment admission issue requires --input ", + }, + )?; Ok(PaymentPlan { action: PaymentAction::IssueAdmission(PaymentAdmissionPlan { - input: input.ok_or_else(|| { - "runx payment admission issue requires --input ".to_owned() - })?, - json, + input: parsed.input.into_payment_source(), + json: true, }), }) } @@ -1080,15 +1035,74 @@ fn parse_parser_plan(args: &[OsString]) -> Result { if subcommand != "eval" { return Err(format!("unknown parser subcommand {subcommand}")); } + let parsed = parse_json_eval_input( + args, + 2, + JsonEvalCommand { + command: "parser", + subject: "parser eval", + duplicate_input: "runx parser eval accepts exactly one --input", + requires_json: "runx parser eval requires --json", + requires_input: "runx parser eval requires --input ", + }, + )?; + Ok(ParserPlan { + input: parsed.input.into_parser_source(), + json: true, + }) +} + +struct JsonEvalCommand { + command: &'static str, + subject: &'static str, + duplicate_input: &'static str, + requires_json: &'static str, + requires_input: &'static str, +} + +struct JsonEvalPlan { + input: JsonEvalInput, +} + +enum JsonEvalInput { + Stdin, + Path(PathBuf), +} + +impl JsonEvalInput { + fn into_kernel_source(self) -> KernelInputSource { + match self { + Self::Stdin => KernelInputSource::Stdin, + Self::Path(path) => KernelInputSource::Path(path), + } + } + + fn into_payment_source(self) -> PaymentInputSource { + match self { + Self::Stdin => PaymentInputSource::Stdin, + Self::Path(path) => PaymentInputSource::Path(path), + } + } + fn into_parser_source(self) -> ParserInputSource { + match self { + Self::Stdin => ParserInputSource::Stdin, + Self::Path(path) => ParserInputSource::Path(path), + } + } +} + +fn parse_json_eval_input( + args: &[OsString], + mut index: usize, + command: JsonEvalCommand, +) -> Result { let mut input = None; let mut json = false; - let mut index = 2; - while index < args.len() { - let token = os_arg(args, index, "parser")?; + let token = os_arg(args, index, command.command)?; if !token.starts_with("--") { - return Err(format!("unexpected parser eval argument {token}")); + return Err(format!("unexpected {} argument {token}", command.subject)); } let (flag, inline_value) = split_flag(token); @@ -1102,27 +1116,25 @@ fn parse_parser_plan(args: &[OsString]) -> Result { } "--input" => { if input.is_some() { - return Err("runx parser eval accepts exactly one --input".to_owned()); + return Err(command.duplicate_input.to_owned()); } - let (value, next_index) = flag_value(args, index, flag, inline_value, "parser")?; + let (value, next_index) = + flag_value(args, index, flag, inline_value, command.command)?; input = Some(if value == "-" { - ParserInputSource::Stdin + JsonEvalInput::Stdin } else { - ParserInputSource::Path(PathBuf::from(value)) + JsonEvalInput::Path(PathBuf::from(value)) }); index = next_index; } - _ => return Err(format!("unknown parser eval flag {flag}")), + _ => return Err(format!("unknown {} flag {flag}", command.subject)), } } - if !json { - return Err("runx parser eval requires --json".to_owned()); + return Err(command.requires_json.to_owned()); } - - Ok(ParserPlan { - input: input.ok_or_else(|| "runx parser eval requires --input ".to_owned())?, - json, + Ok(JsonEvalPlan { + input: input.ok_or_else(|| command.requires_input.to_owned())?, }) } diff --git a/crates/runx-cli/src/login.rs b/crates/runx-cli/src/login.rs index 032ae0c19..21e5e0ac5 100644 --- a/crates/runx-cli/src/login.rs +++ b/crates/runx-cli/src/login.rs @@ -26,6 +26,7 @@ const DEFAULT_LOGIN_TIMEOUT_SECONDS: u64 = 180; pub struct LoginPlan { pub api_base_url: Option, pub provider: Option, + pub purpose: Option, pub allow_local_api: bool, pub json: bool, } @@ -139,6 +140,8 @@ struct LoginStartResponse { struct LoginStartRequest<'a> { #[serde(skip_serializing_if = "Option::is_none")] provider: Option<&'a str>, + #[serde(skip_serializing_if = "Option::is_none")] + purpose: Option<&'a str>, } #[derive(Clone, Debug, serde::Serialize, PartialEq, Eq)] @@ -170,24 +173,25 @@ struct LoginResult { pub fn parse_login_plan(args: &[OsString]) -> Result { let mut api_base_url = None; let mut provider = None; + let mut purpose = None; let mut allow_local_api = false; let mut json = false; let mut index = 1; while index < args.len() { let arg = os_arg(args, index, "login")?; - if !arg.starts_with("--") { + if !arg.starts_with('-') { return Err(format!("unexpected login argument {arg}")); } let (flag, inline_value) = split_flag(arg); match flag { - "--json" => { + "--json" | "-j" => { if inline_value.is_some() { return Err("--json does not take a value".to_owned()); } json = true; index += 1; } - "--api-base-url" | "--apiBaseUrl" => { + "--api-base-url" | "--api-url" | "--apiBaseUrl" => { let (value, next_index) = flag_value(args, index, flag, inline_value, "login")?; api_base_url = Some(value); index = next_index; @@ -197,7 +201,12 @@ pub fn parse_login_plan(args: &[OsString]) -> Result { provider = Some(value); index = next_index; } - "--allow-local-api" | "--allowLocalApi" => { + "--for" | "--purpose" => { + let (value, next_index) = flag_value(args, index, flag, inline_value, "login")?; + purpose = Some(value); + index = next_index; + } + "--allow-local-api" | "--local-api" | "--allowLocalApi" => { if inline_value.is_some() { return Err("--allow-local-api does not take a value".to_owned()); } @@ -210,6 +219,7 @@ pub fn parse_login_plan(args: &[OsString]) -> Result { Ok(LoginPlan { api_base_url, provider, + purpose, allow_local_api, json, }) @@ -264,7 +274,12 @@ fn run_login_command_with_transport( sleep: impl Fn(Duration), ) -> Result { let base_url = resolve_public_api_base_url(plan, env); - let started = start_login_session(transport, &base_url, plan.provider.as_deref())?; + let started = start_login_session( + transport, + &base_url, + plan.provider.as_deref(), + plan.purpose.as_deref(), + )?; let signin_url = started .authorization_url .as_deref() @@ -338,9 +353,11 @@ fn start_login_session( transport: &T, base_url: &str, provider: Option<&str>, + purpose: Option<&str>, ) -> Result { let request = LoginStartRequest { provider: provider.map(str::trim).filter(|value| !value.is_empty()), + purpose: purpose.map(str::trim).filter(|value| !value.is_empty()), }; let response = transport.send(HttpRequest { method: HttpMethod::Post, diff --git a/crates/runx-cli/src/login_tests.rs b/crates/runx-cli/src/login_tests.rs index ac7113532..98e49e687 100644 --- a/crates/runx-cli/src/login_tests.rs +++ b/crates/runx-cli/src/login_tests.rs @@ -33,18 +33,21 @@ impl Transport for StubTransport { fn parses_login_plan() -> Result<(), String> { let args = vec![ OsString::from("login"), - OsString::from("--api-base-url"), + OsString::from("--api-url"), OsString::from("https://runx.test/"), OsString::from("--provider"), OsString::from("github"), - OsString::from("--allow-local-api"), - OsString::from("--json"), + OsString::from("--for"), + OsString::from("publish"), + OsString::from("--local-api"), + OsString::from("-j"), ]; assert_eq!( parse_login_plan(&args)?, LoginPlan { api_base_url: Some("https://runx.test/".to_owned()), provider: Some("github".to_owned()), + purpose: Some("publish".to_owned()), allow_local_api: true, json: true, } @@ -93,6 +96,7 @@ fn login_exchange_stores_encrypted_public_api_token() -> Result<(), Box Result<(), Box Result<(), String> { &LoginPlan { api_base_url: Some("https://runx.test/".to_owned()), provider: Some("bad".to_owned()), + purpose: None, allow_local_api: false, json: false, }, diff --git a/crates/runx-cli/src/main.rs b/crates/runx-cli/src/main.rs index b68cd0fb7..c989b2c95 100644 --- a/crates/runx-cli/src/main.rs +++ b/crates/runx-cli/src/main.rs @@ -249,6 +249,7 @@ fn run_inline_harness(skill_path: &Path, receipt_dir: Option<&OsString>) -> Exit let request = runx_runtime::InlineHarnessRequest { skill_path: skill_path.to_path_buf(), receipt_dir: receipt_dir.map(PathBuf::from), + env: None, }; let report = match runx_cli::runtime::local_orchestrator().run_inline_harness(&request) { Ok(report) => report, diff --git a/crates/runx-cli/src/official_skills.rs b/crates/runx-cli/src/official_skills.rs index 79f7b111e..2d4b4a1d9 100644 --- a/crates/runx-cli/src/official_skills.rs +++ b/crates/runx-cli/src/official_skills.rs @@ -15,6 +15,11 @@ pub(crate) const OFFICIAL_SKILLS: &[OfficialSkillLockEntry] = &[ version: "sha-79c56911c0ba", digest: "54bf1ed1013ebc91a5491fb86f15a1bda2e872ac073a12680c58278af0867528", }, + OfficialSkillLockEntry { + skill_id: "runx/business-ops", + version: "sha-ca97efa3deb0", + digest: "27c05e8e30d8e925c93ecd51e535add38a4ceeabbfc76acf73bb0a578fee3e11", + }, OfficialSkillLockEntry { skill_id: "runx/charge", version: "sha-0e2f6aef60db", @@ -30,6 +35,11 @@ pub(crate) const OFFICIAL_SKILLS: &[OfficialSkillLockEntry] = &[ version: "sha-c2d071df7f50", digest: "08cefe802c15e5be7d32ae9a363a6c42168e86f7fab92890e5ce5c994af367c9", }, + OfficialSkillLockEntry { + skill_id: "runx/dependency-cve-audit", + version: "sha-e9e461e41ea3", + digest: "c19ec9fdeb088daab950b7c2e1f3757880de9702e31e40b57e2f65c0c4033348", + }, OfficialSkillLockEntry { skill_id: "runx/design-skill", version: "sha-0353a69bc33f", @@ -82,7 +92,7 @@ pub(crate) const OFFICIAL_SKILLS: &[OfficialSkillLockEntry] = &[ }, OfficialSkillLockEntry { skill_id: "runx/issue-intake", - version: "sha-94c248e98a1c", + version: "sha-15369469618b", digest: "cc964980fe249ac3633e7b30c664648f0df9406a0254ede9bb0e3cbcdebdd603", }, OfficialSkillLockEntry { @@ -112,7 +122,7 @@ pub(crate) const OFFICIAL_SKILLS: &[OfficialSkillLockEntry] = &[ }, OfficialSkillLockEntry { skill_id: "runx/least-privilege-auditor", - version: "sha-b2913b95c069", + version: "sha-e5c3622556d9", digest: "244df5dd8eed7900d1987c76060893d3c9cd65f420c5b8c177b19fa4e0b81ac2", }, OfficialSkillLockEntry { @@ -192,8 +202,8 @@ pub(crate) const OFFICIAL_SKILLS: &[OfficialSkillLockEntry] = &[ }, OfficialSkillLockEntry { skill_id: "runx/pr-review-note", - version: "sha-dd1608925a76", - digest: "1b9f34f9e7f5355a10babbd154333db4b1b94fa16668583438166b91eec95e0a", + version: "sha-537dd9fc3c6b", + digest: "b073ec884f56c9e412d0c1039d5f28f163df0f5530eb0bee922ed4c557955c52", }, OfficialSkillLockEntry { skill_id: "runx/prior-art", @@ -237,8 +247,8 @@ pub(crate) const OFFICIAL_SKILLS: &[OfficialSkillLockEntry] = &[ }, OfficialSkillLockEntry { skill_id: "runx/review-skill", - version: "sha-a27dde2ab9bb", - digest: "09b4a6ec017f9d75536c6db21c60667bd855a20b0b20f53054f63143cbb9d13d", + version: "sha-622805df5ff3", + digest: "6fc1b341d55e3c6be8a5f7693dfe3312654b89a14f88fe42e4ffc84a65a9cd09", }, OfficialSkillLockEntry { skill_id: "runx/run-history-analyst", @@ -247,8 +257,8 @@ pub(crate) const OFFICIAL_SKILLS: &[OfficialSkillLockEntry] = &[ }, OfficialSkillLockEntry { skill_id: "runx/runx-operator", - version: "sha-da018df211a3", - digest: "1c8d199a5dd0812a09eb4e785bcd9bfd7af67ec0ef3227149bed1de150a47fff", + version: "sha-0fed07a0dc00", + digest: "9de1d9dffb46b6bb14872b66738d5e9b26f271c6f11595c6a685d4c30e539176", }, OfficialSkillLockEntry { skill_id: "runx/sandbox-harden", @@ -277,8 +287,8 @@ pub(crate) const OFFICIAL_SKILLS: &[OfficialSkillLockEntry] = &[ }, OfficialSkillLockEntry { skill_id: "runx/skill-testing", - version: "sha-c31b54a981c8", - digest: "7fc86c62bd493cb374850d7e9fc4faad94adb318fc3b20947aa2d411a741cc75", + version: "sha-9113dacaa62a", + digest: "93f7a0c009e289862fcc9236effdf0ac75197e9eb042a83200720d23d01cb443", }, OfficialSkillLockEntry { skill_id: "runx/slack-notify", @@ -287,8 +297,8 @@ pub(crate) const OFFICIAL_SKILLS: &[OfficialSkillLockEntry] = &[ }, OfficialSkillLockEntry { skill_id: "runx/sourcey", - version: "sha-d025d3a4701e", - digest: "2bdffb5206cbfc2dc619ffead5d26ad192afe0f2836093d782c7901841713006", + version: "sha-2b08f620e0fa", + digest: "4b6316c7fbb323b7d27d304deb8f11cb8f939dc31e0b74349d56f27abf618504", }, OfficialSkillLockEntry { skill_id: "runx/spend", @@ -315,6 +325,16 @@ pub(crate) const OFFICIAL_SKILLS: &[OfficialSkillLockEntry] = &[ version: "sha-e513d3aed77c", digest: "2bfa94189cd3b7084a3b29e1f83de2d0787d28c5f0c962a15bac76155c24d95f", }, + OfficialSkillLockEntry { + skill_id: "runx/structured-extraction", + version: "sha-f14902374e11", + digest: "ebe37921d3b9cb63aa1ca5232a8075607d3b4ad541af4c1bc901ace749ea404f", + }, + OfficialSkillLockEntry { + skill_id: "runx/support-triage-reply", + version: "sha-fce24eb780f9", + digest: "94beeed742142c6a236eb0ae1db5644d3f868362030089332c6ce3b40b1f86bb", + }, OfficialSkillLockEntry { skill_id: "runx/taste-profile", version: "sha-30ae4695f7a2", @@ -347,8 +367,8 @@ pub(crate) const OFFICIAL_SKILLS: &[OfficialSkillLockEntry] = &[ }, OfficialSkillLockEntry { skill_id: "runx/write-harness", - version: "sha-c989640c5604", - digest: "f4fbf60192335baff43a5d50f3702a17f96a42a25d69508f457cf0e396320528", + version: "sha-f69b01f883e0", + digest: "8fbac78e4c760a124c704ce62aa11ecb8b65b165c72a81fd2c1de163c5bb259b", }, OfficialSkillLockEntry { skill_id: "runx/x402-pay", diff --git a/crates/runx-cli/src/public_api.rs b/crates/runx-cli/src/public_api.rs index 008eec06b..00ae838ac 100644 --- a/crates/runx-cli/src/public_api.rs +++ b/crates/runx-cli/src/public_api.rs @@ -3,7 +3,7 @@ use std::collections::BTreeMap; use runx_runtime::registry::{DefaultRuntimeHttpTransport, RuntimeHttpError}; use serde::Deserialize; -pub(crate) const DEFAULT_BASE_URL: &str = "https://runx.ai"; +pub(crate) const DEFAULT_BASE_URL: &str = "https://api.runx.ai"; const BASE_URL_ENV: &str = "RUNX_PUBLIC_API_BASE_URL"; pub(crate) fn resolve_base_url(explicit: Option<&str>, env: &BTreeMap) -> String { @@ -37,6 +37,19 @@ pub(crate) fn parse_error(body: &str) -> Option { serde_json::from_str::(body) .ok() .map(|envelope| envelope.error) + .or_else(|| { + serde_json::from_str::(body) + .ok() + .map(|envelope| match envelope.error { + PlainError::Message(detail) => ErrorPayload { + code: plain_error_code(&detail).to_owned(), + detail, + hint: None, + retry_after_seconds: None, + }, + PlainError::Payload(payload) => payload, + }) + }) } fn normalize_non_empty_base_url(value: &str) -> Option { @@ -62,3 +75,23 @@ pub(crate) struct ErrorPayload { struct ErrorEnvelope { error: ErrorPayload, } + +#[derive(Deserialize)] +struct PlainErrorEnvelope { + error: PlainError, +} + +#[derive(Deserialize)] +#[serde(untagged)] +enum PlainError { + Message(String), + Payload(ErrorPayload), +} + +fn plain_error_code(detail: &str) -> &'static str { + if detail.contains("Missing required scope") { + "missing_scope" + } else { + "api_error" + } +} diff --git a/crates/runx-cli/src/publish.rs b/crates/runx-cli/src/publish.rs index 20e874678..2833ef246 100644 --- a/crates/runx-cli/src/publish.rs +++ b/crates/runx-cli/src/publish.rs @@ -114,10 +114,13 @@ impl fmt::Display for PublishError { "runx-api publish returned invalid JSON: {message}" ) } - Self::RunxApi { code, detail, .. } => { + Self::RunxApi { + code, detail, hint, .. + } => { write!( formatter, - "runx-api publish returned error [{code}]: {detail}" + "{}", + publish_error_message(code, detail, hint.as_deref()) ) } } @@ -132,6 +135,23 @@ impl From for PublishError { } } +fn publish_error_message(code: &str, detail: &str, hint: Option<&str>) -> String { + if code == "missing_scope" && detail.contains("receipts:write") { + return [ + "This token can publish skills but not receipts.", + "The receipt notary requires `receipts:write`.", + "Use `runx publish --token ` or set `RUNX_PUBLIC_API_TOKEN` to a receipt-capable token.", + "Your stored login token is still valid for the scopes it was issued with.", + ] + .join(" "); + } + let mut message = format!("runx-api publish returned error [{code}]: {detail}"); + if let Some(hint) = hint.filter(|value| !value.trim().is_empty()) { + message.push_str(&format!(" Hint: {hint}")); + } + message +} + #[derive(Clone, Debug)] struct PublishOptions<'a> { base_url: &'a str, @@ -165,7 +185,7 @@ pub fn parse_publish_plan(args: &[OsString]) -> Result { let mut index = 1; while index < args.len() { let arg = os_arg(args, index, "publish")?; - if !arg.starts_with("--") { + if !arg.starts_with('-') { if receipt_path.is_some() { return Err(PublishCliError::ExtraReceipt.to_string()); } @@ -175,14 +195,14 @@ pub fn parse_publish_plan(args: &[OsString]) -> Result { } let (flag, inline_value) = split_flag(arg); match flag { - "--json" => { + "--json" | "-j" => { if inline_value.is_some() { return Err("--json does not take a value".to_owned()); } json = true; index += 1; } - "--api-base-url" | "--apiBaseUrl" => { + "--api-base-url" | "--api-url" | "--apiBaseUrl" => { let (value, next_index) = flag_value(args, index, flag, inline_value, "publish")?; api_base_url = Some(value); index = next_index; @@ -192,7 +212,7 @@ pub fn parse_publish_plan(args: &[OsString]) -> Result { token = Some(value); index = next_index; } - "--allow-local-api" | "--allowLocalApi" => { + "--allow-local-api" | "--local-api" | "--allowLocalApi" => { if inline_value.is_some() { return Err("--allow-local-api does not take a value".to_owned()); } diff --git a/crates/runx-cli/src/publish_tests.rs b/crates/runx-cli/src/publish_tests.rs index 86d1c5543..5e59c5b7f 100644 --- a/crates/runx-cli/src/publish_tests.rs +++ b/crates/runx-cli/src/publish_tests.rs @@ -33,11 +33,12 @@ fn parses_publish_plan() -> Result<(), String> { let args = vec![ OsString::from("publish"), OsString::from("receipt.json"), - OsString::from("--api-base-url"), + OsString::from("--api-url"), OsString::from("https://runx.test/"), OsString::from("--token"), OsString::from("rxk_test"), - OsString::from("--json"), + OsString::from("--local-api"), + OsString::from("-j"), ]; let plan = parse_publish_plan(&args)?; assert_eq!( @@ -46,7 +47,7 @@ fn parses_publish_plan() -> Result<(), String> { receipt_path: PathBuf::from("receipt.json"), api_base_url: Some("https://runx.test/".to_owned()), token: Some("rxk_test".to_owned()), - allow_local_api: false, + allow_local_api: true, json: true, } ); @@ -126,7 +127,7 @@ fn resolves_publish_endpoint_and_token_precedence() -> Result<(), Box Result<(), PublishCliError> { Ok(()) } +#[test] +fn publish_error_explains_receipt_scope_mismatch() { + let message = PublishError::RunxApi { + code: "missing_scope".to_owned(), + detail: "Missing required scope: receipts:write.".to_owned(), + hint: None, + retry_after_seconds: None, + } + .to_string(); + + assert!(message.contains("can publish skills but not receipts")); + assert!(message.contains("receipts:write")); + assert!(message.contains("runx publish --token")); +} + fn request_json_body(request: &HttpRequest) -> Result { let body = request .body diff --git a/crates/runx-cli/src/registry/output.rs b/crates/runx-cli/src/registry/output.rs index 02aec80d0..fbedf8374 100644 --- a/crates/runx-cli/src/registry/output.rs +++ b/crates/runx-cli/src/registry/output.rs @@ -80,10 +80,22 @@ pub(super) fn render_search( results.len() ); for result in results { + let category = result + .category + .as_ref() + .map(|category| format!(" category {category}\n")) + .or_else(|| { + result + .source_category + .as_ref() + .map(|category| format!(" source-category {category}\n")) + }) + .unwrap_or_default(); output.push_str(&format!( - " - {}@{}\n digest {}\n trust {}\n install {}\n run {}\n", + " - {}@{}\n{} digest {}\n trust {}\n install {}\n run {}\n", result.skill_id, result.version.as_deref().unwrap_or("unknown"), + category, result .digest .as_deref() @@ -102,9 +114,21 @@ pub(super) fn render_read( registry_ref: &str, skill: &runx_runtime::registry::RegistrySkillDetail, ) -> String { + let category = skill + .category + .as_ref() + .map(|category| format!(" category {category}\n")) + .or_else(|| { + skill + .source_category + .as_ref() + .map(|category| format!(" source category {category}\n")) + }) + .unwrap_or_default(); format!( - "\n registry read {registry_ref}\n source {source}\n skill {}\n version {}\n digest {}\n trust {}\n signed {}\n next {}\n\n", + "\n registry read {registry_ref}\n source {source}\n skill {}\n{} version {}\n digest {}\n trust {}\n signed {}\n next {}\n\n", skill.skill_id, + category, skill.version, digest_label(&skill.digest), trust_tier_label(&skill.trust_tier), diff --git a/crates/runx-cli/src/registry/package.rs b/crates/runx-cli/src/registry/package.rs index eccd75507..d37a8af23 100644 --- a/crates/runx-cli/src/registry/package.rs +++ b/crates/runx-cli/src/registry/package.rs @@ -9,7 +9,10 @@ use std::process; use std::time::{SystemTime, UNIX_EPOCH}; use runx_contracts::{JsonObject, JsonValue}; -use runx_runtime::registry::RegistryPublishHarnessReport; +use runx_runtime::{ + RUNX_RECEIPT_SIGN_ED25519_SEED_BASE64_ENV, RUNX_RECEIPT_SIGN_ISSUER_TYPE_ENV, + RUNX_RECEIPT_SIGN_KID_ENV, registry::RegistryPublishHarnessReport, +}; use serde::Serialize; use super::{RegistryCliError, internal_error}; @@ -89,6 +92,7 @@ pub(super) fn run_publish_harness( let request = runx_runtime::InlineHarnessRequest { skill_path: harness_path.to_path_buf(), receipt_dir: Some(receipt_dir.clone()), + env: Some(publish_harness_env()), }; let report = crate::runtime::local_orchestrator().run_inline_harness(&request); let _ignored = fs::remove_dir_all(&receipt_dir); @@ -132,6 +136,43 @@ struct PublishHarnessPackage { const MAX_REMOTE_PUBLISH_FILE_BYTES: u64 = 512 * 1024; const MAX_REMOTE_PUBLISH_TOTAL_BYTES: u64 = 2 * 1024 * 1024; const MAX_REMOTE_PUBLISH_FILE_COUNT: usize = 128; +const PUBLISH_HARNESS_SIGNING_KID: &str = "runx-publish-harness-local"; +const PUBLISH_HARNESS_SIGNING_SEED_BASE64: &str = "QkJCQkJCQkJCQkJCQkJCQkJCQkJCQkJCQkJCQkJCQkI="; +const PUBLISH_HARNESS_SIGNING_ISSUER_TYPE: &str = "ci"; + +fn publish_harness_env() -> BTreeMap { + let mut env = env::vars().collect(); + ensure_publish_harness_signing_env(&mut env); + env +} + +fn ensure_publish_harness_signing_env(env: &mut BTreeMap) { + if [ + RUNX_RECEIPT_SIGN_KID_ENV, + RUNX_RECEIPT_SIGN_ED25519_SEED_BASE64_ENV, + RUNX_RECEIPT_SIGN_ISSUER_TYPE_ENV, + ] + .iter() + .all(|name| env_value_is_blank(env, name)) + { + env.insert( + RUNX_RECEIPT_SIGN_KID_ENV.to_owned(), + PUBLISH_HARNESS_SIGNING_KID.to_owned(), + ); + env.insert( + RUNX_RECEIPT_SIGN_ED25519_SEED_BASE64_ENV.to_owned(), + PUBLISH_HARNESS_SIGNING_SEED_BASE64.to_owned(), + ); + env.insert( + RUNX_RECEIPT_SIGN_ISSUER_TYPE_ENV.to_owned(), + PUBLISH_HARNESS_SIGNING_ISSUER_TYPE.to_owned(), + ); + } +} + +fn env_value_is_blank(env: &BTreeMap, name: &str) -> bool { + env.get(name).is_none_or(|value| value.trim().is_empty()) +} fn publish_harness_package( markdown: &str, @@ -562,10 +603,55 @@ fn publish_harness_report( #[cfg(test)] mod tests { use super::{ - collect_publish_package_files, should_reject_remote_publish_file, unique_temp_dir, + PUBLISH_HARNESS_SIGNING_ISSUER_TYPE, PUBLISH_HARNESS_SIGNING_KID, + collect_publish_package_files, ensure_publish_harness_signing_env, + should_reject_remote_publish_file, unique_temp_dir, }; use std::fs; + use runx_runtime::{ + RUNX_RECEIPT_SIGN_ED25519_SEED_BASE64_ENV, RUNX_RECEIPT_SIGN_ISSUER_TYPE_ENV, + RUNX_RECEIPT_SIGN_KID_ENV, + }; + + #[test] + fn publish_harness_supplies_local_signing_env_for_fresh_users() { + let mut env = std::collections::BTreeMap::new(); + + ensure_publish_harness_signing_env(&mut env); + + assert_eq!( + env.get(RUNX_RECEIPT_SIGN_KID_ENV).map(String::as_str), + Some(PUBLISH_HARNESS_SIGNING_KID) + ); + assert_eq!( + env.get(RUNX_RECEIPT_SIGN_ISSUER_TYPE_ENV) + .map(String::as_str), + Some(PUBLISH_HARNESS_SIGNING_ISSUER_TYPE) + ); + assert!( + env.get(RUNX_RECEIPT_SIGN_ED25519_SEED_BASE64_ENV) + .is_some_and(|value| !value.trim().is_empty()) + ); + } + + #[test] + fn publish_harness_does_not_mask_partial_signing_env() { + let mut env = std::collections::BTreeMap::from([( + RUNX_RECEIPT_SIGN_KID_ENV.to_owned(), + "explicit-kid".to_owned(), + )]); + + ensure_publish_harness_signing_env(&mut env); + + assert_eq!( + env.get(RUNX_RECEIPT_SIGN_KID_ENV).map(String::as_str), + Some("explicit-kid") + ); + assert!(!env.contains_key(RUNX_RECEIPT_SIGN_ED25519_SEED_BASE64_ENV)); + assert!(!env.contains_key(RUNX_RECEIPT_SIGN_ISSUER_TYPE_ENV)); + } + #[test] fn remote_publish_rejects_common_secret_file_names() { for path in [ diff --git a/crates/runx-cli/src/scaffold.rs b/crates/runx-cli/src/scaffold.rs index 1e3d26bd8..9319a66e5 100644 --- a/crates/runx-cli/src/scaffold.rs +++ b/crates/runx-cli/src/scaffold.rs @@ -24,8 +24,6 @@ pub fn run_native_new(plan: NewPlan) -> ExitCode { let options = RunxNewOptions { name: plan.name, directory, - cli_package_version: scaffold_cli_package_version(), - authoring_package_version: scaffold_authoring_package_version(), }; match scaffold_runx_package(&options) { @@ -89,9 +87,8 @@ fn render_new_result(json: bool, result: &RunxNewResult) -> ExitCode { &NewJsonResult { status: "success", new: NewCommandResult { - action: "package", + action: "skill", name: &result.name, - packet_namespace: &result.packet_namespace, directory: &result.directory, files: &result.files, next_steps: &result.next_steps, @@ -102,8 +99,7 @@ fn render_new_result(json: bool, result: &RunxNewResult) -> ExitCode { write_stdout(&render_key_values( "runx new", &[ - ("package", Some(result.name.clone())), - ("packet_namespace", Some(result.packet_namespace.clone())), + ("skill", Some(result.name.clone())), ("directory", Some(result.directory.display().to_string())), ("files", Some(result.files.len().to_string())), ("next", Some(result.next_steps.join(" && "))), @@ -272,14 +268,6 @@ fn default_home_runx_dir() -> PathBuf { .join(".runx") } -fn scaffold_cli_package_version() -> String { - env::var("RUNX_CLI_PACKAGE_VERSION").unwrap_or_else(|_| "^0.6.2".to_owned()) -} - -fn scaffold_authoring_package_version() -> String { - env::var("RUNX_AUTHORING_PACKAGE_VERSION").unwrap_or_else(|_| "^0.2.0".to_owned()) -} - #[derive(Serialize)] struct NewJsonResult<'a> { status: &'static str, @@ -290,7 +278,6 @@ struct NewJsonResult<'a> { struct NewCommandResult<'a> { action: &'static str, name: &'a str, - packet_namespace: &'a str, directory: &'a Path, files: &'a [String], next_steps: &'a [String], diff --git a/crates/runx-cli/src/skill/parser.rs b/crates/runx-cli/src/skill/parser.rs index 59d98302c..903e4cac3 100644 --- a/crates/runx-cli/src/skill/parser.rs +++ b/crates/runx-cli/src/skill/parser.rs @@ -1,10 +1,15 @@ +// rust-style-allow: large-file - skill CLI parsing keeps shared state and +// option finalization in one module until the native parser surface stabilizes. use std::collections::BTreeMap; use std::env; use std::ffi::OsString; +use std::fs; use std::path::{Path, PathBuf}; use runx_contracts::JsonValue; use runx_runtime::orchestrator::LocalCredentialDescriptor; +use runx_runtime::{resolve_path_from_user_input, resolve_runx_home_dir}; +use serde::Deserialize; use super::SkillPlan; use super::inputs::{parse_direct_input_arg, parse_input_arg}; @@ -18,7 +23,9 @@ pub fn parse_skill_plan(args: &[OsString]) -> Result { index += 1; } - let local_credential = finalize_local_credential(&state)?; + let env = env::vars().collect(); + let cwd = env::current_dir().unwrap_or_else(|_| PathBuf::from(".")); + let local_credential = finalize_local_credential(&state, &env, &cwd)?; let Some(skill_path) = state.skill_path.as_ref() else { return Err("runx skill requires a skill package path".to_owned()); @@ -58,6 +65,7 @@ struct SkillParseState { json: bool, inputs: BTreeMap, credential: Option, + credential_profile: Option, secret_env: Option<(String, String)>, } @@ -68,45 +76,71 @@ struct CredentialBinding { scopes: Vec, } +#[derive(Deserialize)] +#[serde(deny_unknown_fields)] +struct CredentialProfilesFile { + profiles: BTreeMap, +} + +#[derive(Deserialize)] +#[serde(deny_unknown_fields)] +struct CredentialProfile { + credential: String, + secret_env: String, +} + fn parse_credential_binding(value: &str) -> Result { - let mut parts = value.splitn(4, ':'); - let provider = parts - .next() - .filter(|part| !part.is_empty()) - .ok_or_else(|| { - "runx skill --credential requires ::".to_owned() - })?; - let auth_mode = parts - .next() - .filter(|part| !part.is_empty()) - .ok_or_else(|| { - "runx skill --credential requires ::".to_owned() - })?; - let material_ref = parts - .next() - .filter(|part| !part.is_empty()) - .ok_or_else(|| { - "runx skill --credential requires ::".to_owned() - })?; - let scopes = parts - .next() - .map(|raw| { - raw.split(',') - .map(str::trim) - .filter(|scope| !scope.is_empty()) - .map(ToOwned::to_owned) - .collect() - }) - .unwrap_or_default(); + let (provider, rest) = value.split_once(':').ok_or_else(credential_usage_error)?; + let provider = non_empty_credential_part(provider)?; + let (auth_mode, rest) = rest.split_once(':').ok_or_else(credential_usage_error)?; + let auth_mode = non_empty_credential_part(auth_mode)?; + let (material_ref, scopes) = split_material_ref_and_scopes(rest)?; Ok(CredentialBinding { provider: provider.to_owned(), auth_mode: auth_mode.to_owned(), - material_ref: material_ref.to_owned(), + material_ref, scopes, }) } +fn split_material_ref_and_scopes(value: &str) -> Result<(String, Vec), String> { + let value = non_empty_credential_part(value)?; + let Some(index) = value.rfind(':') else { + return Ok((value.to_owned(), Vec::new())); + }; + if value[index..].starts_with("://") { + return Ok((value.to_owned(), Vec::new())); + } + let material_ref = non_empty_credential_part(&value[..index])?.to_owned(); + let scopes = value[index + 1..] + .split(',') + .map(str::trim) + .filter(|scope| !scope.is_empty()) + .map(ToOwned::to_owned) + .collect(); + Ok((material_ref, scopes)) +} + +fn non_empty_credential_part(value: &str) -> Result<&str, String> { + let value = value.trim(); + if value.is_empty() { + return Err(credential_usage_error()); + } + Ok(value) +} + +fn credential_usage_error() -> String { + "runx skill --credential requires ::".to_owned() +} + fn parse_secret_env(value: &str) -> Result<(String, String), String> { + parse_secret_env_from(value, |name| env::var(name).ok()) +} + +fn parse_secret_env_from( + value: &str, + lookup: impl Fn(&str) -> Option, +) -> Result<(String, String), String> { if value.contains('=') { return Err( "runx skill --secret-env accepts an environment variable name, not an inline value" @@ -117,8 +151,8 @@ fn parse_secret_env(value: &str) -> Result<(String, String), String> { if name.is_empty() { return Err("runx skill --secret-env requires a non-empty env var name".to_owned()); } - let secret = env::var(name) - .map_err(|_| format!("runx skill --secret-env env var '{name}' is not set"))?; + let secret = lookup(name) + .ok_or_else(|| format!("runx skill --secret-env env var '{name}' is not set"))?; if secret.trim().is_empty() { return Err("runx skill --secret-env requires a non-empty secret value".to_owned()); } @@ -127,7 +161,21 @@ fn parse_secret_env(value: &str) -> Result<(String, String), String> { fn finalize_local_credential( state: &SkillParseState, + env: &BTreeMap, + cwd: &Path, ) -> Result, String> { + if let Some(profile) = state.credential_profile.as_ref() { + if state.credential.is_some() || state.secret_env.is_some() { + return Err( + "runx skill --credential-profile cannot be combined with --credential or --secret-env" + .to_owned(), + ); + } + let profile = load_credential_profile(profile, env, cwd)?; + let credential = parse_credential_binding(&profile.credential)?; + let secret_env = parse_secret_env_from(&profile.secret_env, |name| env.get(name).cloned())?; + return Ok(Some(local_credential_descriptor(&credential, &secret_env))); + } match (&state.credential, &state.secret_env) { (None, None) => Ok(None), (Some(_), None) => { @@ -138,16 +186,100 @@ fn finalize_local_credential( "runx skill --secret-env requires --credential ::" .to_owned() })?; - Ok(Some(LocalCredentialDescriptor { - provider: binding.provider.clone(), - auth_mode: binding.auth_mode.clone(), - env_var: env_var.clone(), - material_ref: binding.material_ref.clone(), - scopes: binding.scopes.clone(), - secret: secret.clone(), - })) + Ok(Some(local_credential_descriptor( + binding, + &(env_var.clone(), secret.clone()), + ))) + } + } +} + +fn local_credential_descriptor( + binding: &CredentialBinding, + secret_env: &(String, String), +) -> LocalCredentialDescriptor { + LocalCredentialDescriptor { + provider: binding.provider.clone(), + auth_mode: binding.auth_mode.clone(), + env_var: secret_env.0.clone(), + material_ref: binding.material_ref.clone(), + scopes: binding.scopes.clone(), + secret: secret_env.1.clone(), + } +} + +fn load_credential_profile( + profile_name: &str, + env: &BTreeMap, + cwd: &Path, +) -> Result { + let profile_name = profile_name.trim(); + if profile_name.is_empty() { + return Err("runx skill --credential-profile requires a non-empty name".to_owned()); + } + let paths = credential_profile_paths(env, cwd); + for path in &paths { + let contents = match fs::read_to_string(path) { + Ok(contents) => contents, + Err(error) if error.kind() == std::io::ErrorKind::NotFound => continue, + Err(error) => { + return Err(format!( + "runx skill could not read credential profile file {}: {error}", + path.display() + )); + } + }; + let parsed: CredentialProfilesFile = serde_json::from_str(&contents).map_err(|error| { + format!( + "runx skill credential profile file {} is invalid JSON: {error}", + path.display() + ) + })?; + if let Some(profile) = parsed.profiles.into_iter().find_map(|(name, profile)| { + if name == profile_name { + Some(profile) + } else { + None + } + }) { + return Ok(profile); } } + let searched = paths + .iter() + .map(|path| path.display().to_string()) + .collect::>() + .join(", "); + Err(format!( + "runx skill credential profile '{profile_name}' was not found; searched {searched}" + )) +} + +fn credential_profile_paths(env: &BTreeMap, cwd: &Path) -> Vec { + if let Some(path) = env + .get("RUNX_CREDENTIAL_PROFILES") + .filter(|value| !value.trim().is_empty()) + { + return vec![resolve_path_from_user_input(path, env, cwd, true)]; + } + let mut paths = Vec::new(); + if let Some(project_dir) = env + .get("RUNX_PROJECT_DIR") + .filter(|value| !value.trim().is_empty()) + .map(|value| resolve_path_from_user_input(value, env, cwd, true)) + .or_else(|| nearest_project_runx_dir(cwd)) + { + paths.push(project_dir.join("credentials.json")); + } + paths.push(resolve_runx_home_dir(env, cwd).join("credentials.json")); + paths.dedup(); + paths +} + +fn nearest_project_runx_dir(cwd: &Path) -> Option { + cwd.ancestors() + .map(|ancestor| ancestor.join(".runx")) + .find(|candidate| candidate.is_dir()) } fn reject_resolver_flags_for_skill_management_action( @@ -191,10 +323,24 @@ fn parse_skill_arg( value if value.starts_with("--receipt-dir=") => { state.receipt_dir = Some(PathBuf::from(value.trim_start_matches("--receipt-dir="))); } + value if value.starts_with("-R=") => { + state.receipt_dir = Some(PathBuf::from(value.trim_start_matches("-R="))); + } + value if value.starts_with("--receipts=") => { + state.receipt_dir = Some(PathBuf::from(value.trim_start_matches("--receipts="))); + } "--receipt-dir" => { index += 1; state.receipt_dir = Some(PathBuf::from(string_arg(args, index)?)); } + "--receipts" => { + index += 1; + state.receipt_dir = Some(PathBuf::from(string_arg(args, index)?)); + } + "-R" => { + index += 1; + state.receipt_dir = Some(PathBuf::from(string_arg(args, index)?)); + } value if value.starts_with("--run-id=") => { state.run_id = Some(value.trim_start_matches("--run-id=").to_owned()); } @@ -251,7 +397,16 @@ fn parse_skill_arg( &mut state.inputs, )?; } + value if value.starts_with("-i=") => { + index = parse_input_arg( + args, + index, + Some(value.trim_start_matches("-i=")), + &mut state.inputs, + )?; + } "--input" => index = parse_input_arg(args, index, None, &mut state.inputs)?, + "-i" => index = parse_input_arg(args, index, None, &mut state.inputs)?, value if value.starts_with("--credential=") => { state.credential = Some(parse_credential_binding( value.trim_start_matches("--credential="), @@ -261,6 +416,32 @@ fn parse_skill_arg( index += 1; state.credential = Some(parse_credential_binding(&string_arg(args, index)?)?); } + value if value.starts_with("--credential-profile=") => { + state.credential_profile = Some(non_empty_flag_value( + "--credential-profile", + value.trim_start_matches("--credential-profile="), + )?); + } + value if value.starts_with("--profile=") => { + state.credential_profile = Some(non_empty_flag_value( + "--profile", + value.trim_start_matches("--profile="), + )?); + } + "--credential-profile" => { + index += 1; + state.credential_profile = Some(non_empty_flag_value( + "--credential-profile", + &string_arg(args, index)?, + )?); + } + "--profile" | "-p" => { + index += 1; + state.credential_profile = Some(non_empty_flag_value( + "--profile", + &string_arg(args, index)?, + )?); + } value if value.starts_with("--secret-env=") => { state.secret_env = Some(parse_secret_env(value.trim_start_matches("--secret-env="))?); } @@ -268,7 +449,7 @@ fn parse_skill_arg( index += 1; state.secret_env = Some(parse_secret_env(&string_arg(args, index)?)?); } - "--json" => state.json = true, + "--json" | "-j" => state.json = true, "--non-interactive" => {} value if value.starts_with("--") => { index = parse_direct_input_arg(args, index, value, &mut state.inputs)?; @@ -312,3 +493,127 @@ fn string_arg(args: &[OsString], index: usize) -> Result { .map(ToOwned::to_owned) .ok_or_else(|| "runx skill arguments must be UTF-8".to_owned()) } + +#[cfg(test)] +mod tests { + use std::fs; + use std::time::{SystemTime, UNIX_EPOCH}; + + use super::{SkillParseState, finalize_local_credential}; + + #[test] + fn credential_profile_resolves_project_descriptor_and_env_secret() -> Result<(), String> { + let root = unique_temp_dir("runx-credential-profile")?; + let runx_dir = root.join(".runx"); + fs::create_dir_all(&runx_dir).map_err(|error| error.to_string())?; + fs::write( + runx_dir.join("credentials.json"), + r#"{ + "profiles": { + "example": { + "credential": "example:bearer:local://example/internal:example.review", + "secret_env": "INTERNAL_SYNC_SECRET" + } + } +} +"#, + ) + .map_err(|error| error.to_string())?; + + let state = SkillParseState { + credential_profile: Some("example".to_owned()), + ..Default::default() + }; + let env = [ + ( + "RUNX_PROJECT_DIR".to_owned(), + runx_dir.to_string_lossy().into_owned(), + ), + ("INTERNAL_SYNC_SECRET".to_owned(), "secret-value".to_owned()), + ] + .into_iter() + .collect(); + let credential = finalize_local_credential(&state, &env, &root)? + .ok_or_else(|| "credential profile did not resolve".to_owned())?; + + assert_eq!(credential.provider, "example"); + assert_eq!(credential.auth_mode, "bearer"); + assert_eq!(credential.material_ref, "local://example/internal"); + assert_eq!(credential.env_var, "INTERNAL_SYNC_SECRET"); + assert_eq!(credential.secret, "secret-value"); + assert_eq!(credential.scopes, vec!["example.review"]); + + fs::remove_dir_all(root).map_err(|error| error.to_string())?; + Ok(()) + } + + #[test] + fn credential_profile_rejects_manual_credential_flags() -> Result<(), String> { + let state = SkillParseState { + credential_profile: Some("example".to_owned()), + secret_env: Some(("TOKEN".to_owned(), "secret".to_owned())), + ..Default::default() + }; + let error = finalize_local_credential(&state, &Default::default(), &std::env::temp_dir()) + .err() + .ok_or_else(|| "profile unexpectedly combined with manual flags".to_owned())?; + assert!(error.contains("cannot be combined")); + Ok(()) + } + + #[test] + fn short_human_flags_parse_for_skill_runs() -> Result<(), String> { + let args = [ + "skill", + "-p", + "operator", + "-i", + "claim=abc", + "-R", + "receipts", + "-j", + ] + .into_iter() + .map(std::ffi::OsString::from) + .collect::>(); + let mut state = SkillParseState::default(); + let mut index = 1; + while index < args.len() { + index = super::parse_skill_arg(&args, index, &mut state)?; + index += 1; + } + assert_eq!(state.credential_profile.as_deref(), Some("operator")); + assert_eq!( + state.inputs.get("claim"), + Some(&runx_contracts::JsonValue::String("abc".to_owned())) + ); + assert_eq!( + state.receipt_dir.as_deref(), + Some(std::path::Path::new("receipts")) + ); + assert!(state.json); + Ok(()) + } + + #[test] + fn credential_parser_keeps_uri_material_ref_intact() -> Result<(), String> { + let binding = super::parse_credential_binding( + "frantic:bearer:local://frantic/internal:frantic.review,frantic.write", + )?; + assert_eq!(binding.provider, "frantic"); + assert_eq!(binding.auth_mode, "bearer"); + assert_eq!(binding.material_ref, "local://frantic/internal"); + assert_eq!(binding.scopes, vec!["frantic.review", "frantic.write"]); + Ok(()) + } + + fn unique_temp_dir(name: &str) -> Result { + let nanos = SystemTime::now() + .duration_since(UNIX_EPOCH) + .map_err(|error| error.to_string())? + .as_nanos(); + let path = std::env::temp_dir().join(format!("{name}-{}-{nanos}", std::process::id())); + fs::create_dir_all(&path).map_err(|error| error.to_string())?; + Ok(path) + } +} diff --git a/crates/runx-cli/src/url_add.rs b/crates/runx-cli/src/url_add.rs index 6df0f3fea..a473019f9 100644 --- a/crates/runx-cli/src/url_add.rs +++ b/crates/runx-cli/src/url_add.rs @@ -256,7 +256,7 @@ mod tests { let empty_env: BTreeMap = BTreeMap::new(); assert_eq!( resolve_public_api_base_url(&plan_no_override, &empty_env), - "https://runx.ai", + "https://api.runx.ai", ); } diff --git a/crates/runx-cli/src/verify.rs b/crates/runx-cli/src/verify.rs index 76d8b04bc..f977b693b 100644 --- a/crates/runx-cli/src/verify.rs +++ b/crates/runx-cli/src/verify.rs @@ -253,7 +253,7 @@ fn parse_verify_args(args: &[OsString]) -> Result parsed.json = true, + "--json" | "-j" => parsed.json = true, "--allow-local-development-signatures" => { parsed.allow_local_development_signatures = true; } diff --git a/crates/runx-cli/tests/launcher.rs b/crates/runx-cli/tests/launcher.rs index eff6533bd..e4b85120d 100644 --- a/crates/runx-cli/tests/launcher.rs +++ b/crates/runx-cli/tests/launcher.rs @@ -29,11 +29,11 @@ fn top_level_help_and_version_are_native() { let help = help_text(); assert_help_line( &help, - "runx verify [receipt-id] [--receipt-dir dir] [--receipt ] [--notary --notary-key trusted.pem] [--json]", + "runx verify [receipt-id] [--receipt-dir dir] [--receipt ] [--notary --notary-key trusted.pem] [-j|--json]", ); assert_help_line( &help, - "runx skill [--registry url|path] [--digest sha256] [--input key=value] [--runner name] [--flag value] [--receipt-dir dir] [--run-id id --answers file] [--json]", + "runx skill [-p profile] [-i key=value] [-j] [--runner name] [--registry url|path] [--digest sha256] [--flag value] [--credential descriptor --secret-env NAME] [-R dir] [--run-id id --answers file]", ); assert_help_line( &help, @@ -42,7 +42,7 @@ fn top_level_help_and_version_are_native() { assert_help_line(&help, "runx parser eval --input --json"); assert_help_line( &help, - "runx harness [--receipt-dir dir] [--json]", + "runx harness [-R dir] [-j|--json]", ); assert_help_line(&help, "runx doctor [path|authority|registry] [--json]"); assert_help_line( @@ -51,7 +51,7 @@ fn top_level_help_and_version_are_native() { ); assert_help_line( &help, - "runx login [--provider github|google|gitlab] [--api-base-url url] [--allow-local-api] [--json]", + "runx login [--provider github|google|gitlab] [--for default|publish] [--api-url url] [--local-api] [-j|--json]", ); assert!( !help.contains("runx connect"), @@ -98,7 +98,11 @@ fn nested_skill_history_verify_and_publish_help_are_native() { assert_help_line( &skill_help_text(), - "runx skill [--registry url|path] [--digest sha256] [--input key=value] [--runner name] [--flag value] [--receipt-dir dir] [--run-id id --answers file] [--json]", + "runx skill [-p profile] [-i key=value] [-j] [--runner name] [--registry url|path] [--digest sha256] [--flag value] [--credential descriptor --secret-env NAME] [-R dir] [--run-id id --answers file]", + ); + assert_help_line( + &skill_help_text(), + "-p, --profile name Use a local credential profile from .runx/credentials.json", ); assert_help_line( &history_help_text(), @@ -106,11 +110,11 @@ fn nested_skill_history_verify_and_publish_help_are_native() { ); assert_help_line( &verify_help_text(), - "runx verify [receipt-id] [--receipt-dir dir] [--receipt ] [--notary --notary-key trusted.pem] [--json]", + "runx verify [receipt-id] [--receipt-dir dir] [--receipt ] [--notary --notary-key trusted.pem] [-j|--json]", ); assert_help_line( &publish_help_text(), - "runx publish [--api-base-url url] [--token token] [--allow-local-api] [--json]", + "runx publish [--api-url url] [--token token] [--local-api] [-j|--json]", ); } @@ -121,12 +125,15 @@ fn routes_login_to_native_plan() { "login", "--provider", "github", + "--for", + "publish", "--api-base-url", "https://runx.test", "--json", ]), LauncherAction::RunLogin(LoginPlan { provider: Some("github".to_owned()), + purpose: Some("publish".to_owned()), api_base_url: Some("https://runx.test".to_owned()), allow_local_api: false, json: true, @@ -232,10 +239,16 @@ fn mcp_rejects_unknown_shapes_without_delegating() { #[test] fn routes_harness_to_native_runner() { assert_eq!( - plan(&["harness", "fixtures/harness/echo-skill.yaml", "--json"]), + plan(&[ + "harness", + "fixtures/harness/echo-skill.yaml", + "-R", + ".runx/receipts", + "-j" + ]), LauncherAction::RunHarness(HarnessPlan { fixture_paths: vec!["fixtures/harness/echo-skill.yaml".into()], - receipt_dir: None, + receipt_dir: Some(".runx/receipts".into()), }) ); } @@ -692,10 +705,10 @@ fn routes_add_to_native_plan() { #[test] fn routes_tool_to_native_plan_and_rejects_unknown_subcommands() { assert_eq!( - plan(&["tool", "build", "tools/docs/echo", "--json"]), + plan(&["tool", "build", "tools/fixture/minimal", "--json"]), LauncherAction::RunTool(ToolPlan { action: ToolAction::Build, - path: Some(PathBuf::from("tools/docs/echo")), + path: Some(PathBuf::from("tools/fixture/minimal")), ref_or_query: None, all: false, source: None, diff --git a/crates/runx-cli/tests/tool.rs b/crates/runx-cli/tests/tool.rs index a5bedf21c..774ccdd26 100644 --- a/crates/runx-cli/tests/tool.rs +++ b/crates/runx-cli/tests/tool.rs @@ -45,10 +45,10 @@ fn tool_inspect_fixture_catalog_json() -> Result<(), Box> } #[test] -fn tool_build_scaffold_manifest_json() -> Result<(), Box> { - let temp_root = copy_scaffold_fixture("cli_tool_build")?; +fn tool_build_minimal_manifest_json() -> Result<(), Box> { + let temp_root = copy_tool_catalog_build_fixture("cli_tool_build", "minimal")?; let output = runx_command() - .args(["tool", "build", "tools/docs/echo", "--json"]) + .args(["tool", "build", "tools/fixture/minimal", "--json"]) .env("RUNX_CWD", &temp_root) .output()?; @@ -199,18 +199,6 @@ fn optional_oracle_contents(name: &str) -> Result, Box Result> { - let source = repo_root()?.join("fixtures/scaffold/new-docs-demo/files"); - let target = std::env::temp_dir() - .join("runx-tool-cli-tests") - .join(format!("{name}-{}", std::process::id())); - if target.exists() { - fs::remove_dir_all(&target)?; - } - copy_dir(&source, &target)?; - Ok(target) -} - fn copy_tool_catalog_build_fixture( name: &str, fixture_name: &str, diff --git a/crates/runx-contracts/src/lib.rs b/crates/runx-contracts/src/lib.rs index cd3a36628..2a0577dbc 100644 --- a/crates/runx-contracts/src/lib.rs +++ b/crates/runx-contracts/src/lib.rs @@ -157,8 +157,7 @@ pub use operational_proposal::{ OPERATIONAL_PROPOSAL_SCHEMA, OperationalProposal, OperationalProposalAuthority, OperationalProposalHumanGate, OperationalProposalIdempotency, OperationalProposalOutcome, OperationalProposalRecommendedAction, OperationalProposalRedactionStatus, - OperationalProposalReference, OperationalProposalReferenceLink, - OperationalProposalReferenceType, OperationalProposalSchema, + OperationalProposalSchema, }; pub use output::{Output, OutputField, OutputFieldSpec, OutputType}; pub use packet_index::{PacketIndex, PacketIndexEntry, PacketIndexSchema}; diff --git a/crates/runx-contracts/src/operational_proposal.rs b/crates/runx-contracts/src/operational_proposal.rs index 5f8251df5..e1e22f970 100644 --- a/crates/runx-contracts/src/operational_proposal.rs +++ b/crates/runx-contracts/src/operational_proposal.rs @@ -7,7 +7,7 @@ use serde::{Deserialize, Serialize}; use serde_json::{Value, json}; use crate::schema::{Identity, IsoDateTime, NonEmptyString, Property, RunxSchema, object_schema}; -use crate::{JsonObject, ProofKind}; +use crate::{JsonObject, Reference, ReferenceLink}; pub const OPERATIONAL_PROPOSAL_SCHEMA: &str = "runx.operational_proposal.v1"; @@ -25,80 +25,6 @@ pub enum OperationalProposalRedactionStatus { Blocked, } -#[derive(Clone, Copy, Debug, Eq, PartialEq, Serialize, Deserialize, RunxSchema)] -#[serde(rename_all = "snake_case")] -pub enum OperationalProposalReferenceType { - ProviderThread, - ProviderEvent, - ProviderComment, - TrackingItem, - ChangeRequest, - Repository, - SupportTicket, - Signal, - Act, - Receipt, - GraphReceipt, - Artifact, - Verification, - Harness, - Host, - Deployment, - Surface, - Target, - Opportunity, - ThesisAssessment, - Selection, - SkillBinding, - TargetTransitionEntry, - SelectionCycle, - Decision, - ReflectionEntry, - FeedEntry, - Principal, - AuthorityProof, - ScopeAdmission, - Grant, - Mandate, - Credential, - WebhookDelivery, - RedactionPolicy, - ExternalUrl, -} - -/// Provider-neutral reference shape for operational proposal packets. -/// -/// GitHub, Slack, Sentry, and similar systems remain adapters/providers. Their -/// concrete names belong in `provider`, `locator`, and `uri`, not in the -/// shared reference `type` vocabulary used by proposals. -#[derive(Clone, Debug, PartialEq, Eq, Serialize, Deserialize, RunxSchema)] -#[serde(deny_unknown_fields)] -#[runx_schema(id = "runx.operational_proposal.reference.v1")] -pub struct OperationalProposalReference { - #[serde(rename = "type")] - pub reference_type: OperationalProposalReferenceType, - pub uri: NonEmptyString, - #[serde(skip_serializing_if = "Option::is_none")] - pub provider: Option, - #[serde(skip_serializing_if = "Option::is_none")] - pub locator: Option, - #[serde(skip_serializing_if = "Option::is_none")] - pub label: Option, - #[serde(skip_serializing_if = "Option::is_none")] - pub observed_at: Option, - #[serde(skip_serializing_if = "Option::is_none")] - pub proof_kind: Option, -} - -#[derive(Clone, Debug, PartialEq, Eq, Serialize, Deserialize, RunxSchema)] -#[serde(deny_unknown_fields)] -#[runx_schema(id = "runx.operational_proposal.reference_link.v1")] -pub struct OperationalProposalReferenceLink { - pub role: NonEmptyString, - #[serde(rename = "ref")] - pub reference: OperationalProposalReference, -} - #[derive(Clone, Debug, PartialEq, Serialize, Deserialize, RunxSchema)] #[serde(deny_unknown_fields)] pub struct OperationalProposalRecommendedAction { @@ -106,7 +32,7 @@ pub struct OperationalProposalRecommendedAction { pub summary: NonEmptyString, pub mutating: bool, #[serde(default, skip_serializing_if = "Vec::is_empty")] - pub target_refs: Vec, + pub target_refs: Vec, } #[derive(Clone, Debug, PartialEq, Eq, Serialize, Deserialize, RunxSchema)] @@ -166,7 +92,7 @@ pub struct OperationalProposalOutcome { #[serde(skip_serializing_if = "Option::is_none")] pub observed_at: Option, #[serde(default, skip_serializing_if = "Vec::is_empty")] - pub refs: Vec, + pub refs: Vec, } #[derive(Clone, Debug, PartialEq, Serialize, Deserialize)] @@ -177,27 +103,27 @@ pub struct OperationalProposal { pub proposal_kind: NonEmptyString, pub source_event_id: NonEmptyString, pub idempotency: OperationalProposalIdempotency, - pub source_ref: OperationalProposalReference, + pub source_ref: Reference, #[serde(skip_serializing_if = "Option::is_none")] - pub source_thread_ref: Option, - pub hydrated_context_ref: OperationalProposalReference, + pub source_thread_ref: Option, + pub hydrated_context_ref: Reference, pub redaction_status: OperationalProposalRedactionStatus, pub decision_summary: NonEmptyString, pub rationale: NonEmptyString, #[serde(default)] pub recommended_actions: Vec, #[serde(default)] - pub evidence_refs: Vec, + pub evidence_refs: Vec, #[serde(default)] - pub artifact_refs: Vec, + pub artifact_refs: Vec, #[serde(default)] - pub receipt_refs: Vec, + pub receipt_refs: Vec, #[serde(default)] - pub story_refs: Vec, + pub story_refs: Vec, #[serde(default)] - pub result_refs: Vec, + pub result_refs: Vec, #[serde(default)] - pub publication_refs: Vec, + pub publication_refs: Vec, pub owner_route_id: NonEmptyString, pub confidence: f64, #[serde(default)] @@ -236,21 +162,9 @@ impl RunxSchema for OperationalProposal { OperationalProposalIdempotency::json_schema(), true, ), - Property::new( - "source_ref", - OperationalProposalReference::json_schema(), - true, - ), - Property::new( - "source_thread_ref", - OperationalProposalReference::json_schema(), - false, - ), - Property::new( - "hydrated_context_ref", - OperationalProposalReference::json_schema(), - true, - ), + Property::new("source_ref", Reference::json_schema(), true), + Property::new("source_thread_ref", Reference::json_schema(), false), + Property::new("hydrated_context_ref", Reference::json_schema(), true), Property::new( "redaction_status", OperationalProposalRedactionStatus::json_schema(), @@ -263,34 +177,14 @@ impl RunxSchema for OperationalProposal { Vec::::json_schema(), false, ), - Property::new( - "evidence_refs", - Vec::::json_schema(), - false, - ), - Property::new( - "artifact_refs", - Vec::::json_schema(), - false, - ), - Property::new( - "receipt_refs", - Vec::::json_schema(), - false, - ), - Property::new( - "story_refs", - Vec::::json_schema(), - false, - ), - Property::new( - "result_refs", - Vec::::json_schema(), - false, - ), + Property::new("evidence_refs", Vec::::json_schema(), false), + Property::new("artifact_refs", Vec::::json_schema(), false), + Property::new("receipt_refs", Vec::::json_schema(), false), + Property::new("story_refs", Vec::::json_schema(), false), + Property::new("result_refs", Vec::::json_schema(), false), Property::new( "publication_refs", - Vec::::json_schema(), + Vec::::json_schema(), false, ), Property::new("owner_route_id", id_schema(), true), diff --git a/crates/runx-contracts/tests/operational_proposal_fixtures.rs b/crates/runx-contracts/tests/operational_proposal_fixtures.rs index dedc60161..ea3302e16 100644 --- a/crates/runx-contracts/tests/operational_proposal_fixtures.rs +++ b/crates/runx-contracts/tests/operational_proposal_fixtures.rs @@ -19,9 +19,6 @@ const INVALID_FIXTURES: &[&str] = &[ include_str!( "../../../fixtures/contracts/operational-proposal/invalid-product-specific-field.json" ), - include_str!( - "../../../fixtures/contracts/operational-proposal/invalid-provider-locked-reference-type.json" - ), ]; #[derive(Debug, Deserialize)] diff --git a/crates/runx-parser/src/graph/step.rs b/crates/runx-parser/src/graph/step.rs index adf7dd7fc..6a82413a6 100644 --- a/crates/runx-parser/src/graph/step.rs +++ b/crates/runx-parser/src/graph/step.rs @@ -1,3 +1,5 @@ +// rust-style-allow: large-file - graph step validation is kept together so +// field-level diagnostics stay consistent across graph target variants. use std::collections::{BTreeMap, BTreeSet}; use runx_contracts::{JsonObject, JsonValue}; @@ -37,6 +39,10 @@ pub fn validate_step( .unwrap_or_default(); validate_context_skills(&context_skills, field, &target)?; + let inputs = + optional_object(raw_step.get("inputs"), &format!("{field}.inputs"))?.unwrap_or_default(); + reject_step_output_refs_in_inputs(&inputs, previous_step_ids, &format!("{field}.inputs"))?; + Ok(GraphStep { id, label: optional_non_empty_string(raw_step.get("label"), &format!("{field}.label"))?, @@ -49,8 +55,7 @@ pub fn validate_step( )?, artifacts: optional_object(raw_step.get("artifacts"), &format!("{field}.artifacts"))?, runner, - inputs: optional_object(raw_step.get("inputs"), &format!("{field}.inputs"))? - .unwrap_or_default(), + inputs, context_edges: context_edges(&context, previous_step_ids, field)?, context, context_skills, @@ -73,6 +78,65 @@ pub fn validate_step( }) } +fn reject_step_output_refs_in_inputs( + inputs: &JsonObject, + previous_step_ids: &BTreeSet, + field: &str, +) -> Result<(), ValidationError> { + for (key, value) in inputs { + reject_step_output_refs_in_input_value( + value, + previous_step_ids, + &format!("{field}.{key}"), + )?; + } + Ok(()) +} + +fn reject_step_output_refs_in_input_value( + value: &JsonValue, + previous_step_ids: &BTreeSet, + field: &str, +) -> Result<(), ValidationError> { + match value { + JsonValue::String(value) + if looks_like_previous_step_output_ref(value, previous_step_ids) => + { + return Err(validation_error(format!( + "{field} looks like step output reference {value:?}; move it to context if you meant to read a previous step output." + ))); + } + JsonValue::String(_) => {} + JsonValue::Object(object) => { + for (key, value) in object { + reject_step_output_refs_in_input_value( + value, + previous_step_ids, + &format!("{field}.{key}"), + )?; + } + } + JsonValue::Array(values) => { + for (index, value) in values.iter().enumerate() { + reject_step_output_refs_in_input_value( + value, + previous_step_ids, + &format!("{field}.{index}"), + )?; + } + } + _ => {} + } + Ok(()) +} + +fn looks_like_previous_step_output_ref(value: &str, previous_step_ids: &BTreeSet) -> bool { + let Some((from_step, output)) = value.split_once('.') else { + return false; + }; + !output.is_empty() && previous_step_ids.contains(from_step) +} + fn validate_when( value: Option<&JsonValue>, field: &str, diff --git a/crates/runx-parser/src/graph/validate.rs b/crates/runx-parser/src/graph/validate.rs index e6b252c2e..388386777 100644 --- a/crates/runx-parser/src/graph/validate.rs +++ b/crates/runx-parser/src/graph/validate.rs @@ -69,3 +69,54 @@ fn reject_unsupported_top_level(document: &JsonObject) -> Result<(), ValidationE } Ok(()) } + +#[cfg(test)] +mod tests { + use super::{parse_graph_yaml, validate_graph}; + + #[test] + fn inputs_reject_previous_step_output_references() -> Result<(), String> { + let raw = parse_graph_yaml( + r#" +name: bad-input-ref +steps: + - id: select + run: + type: agent-task + - id: review + run: + type: agent-task + inputs: + bounty: select.result +"#, + ) + .map_err(|error| error.to_string())?; + let error = validate_graph(raw) + .err() + .ok_or_else(|| "graph unexpectedly validated".to_owned())?; + let message = error.to_string(); + assert!(message.contains("steps.1.inputs.bounty")); + assert!(message.contains("move it to context")); + Ok(()) + } + + #[test] + fn inputs_allow_literals_that_are_not_previous_step_refs() -> Result<(), String> { + let raw = parse_graph_yaml( + r#" +name: literal-input +steps: + - id: review + run: + type: agent-task + inputs: + literal: select.result + variable: $input.claim + url: https://example.com/a.b +"#, + ) + .map_err(|error| error.to_string())?; + validate_graph(raw).map_err(|error| error.to_string())?; + Ok(()) + } +} diff --git a/crates/runx-parser/src/json_fields.rs b/crates/runx-parser/src/json_fields.rs index 54f901f3b..2bdf245e0 100644 --- a/crates/runx-parser/src/json_fields.rs +++ b/crates/runx-parser/src/json_fields.rs @@ -151,6 +151,23 @@ impl JsonFieldReader { Some(_) => Err(self.validation_error(format!("{field} must be a finite number."))), } } + + pub(crate) fn reject_unknown_fields( + &self, + object: &JsonObject, + field: &str, + allowed: &[&str], + ) -> Result<(), ValidationError> { + for key in object.keys() { + if !allowed.contains(&key.as_str()) { + return Err(self.validation_error(format!( + "{field}.{key} is not supported; allowed fields: {}.", + allowed.join(", ") + ))); + } + } + Ok(()) + } } pub(crate) fn first_value<'a>( diff --git a/crates/runx-parser/src/lib.rs b/crates/runx-parser/src/lib.rs index 7630a0651..d40e5af2a 100644 --- a/crates/runx-parser/src/lib.rs +++ b/crates/runx-parser/src/lib.rs @@ -37,6 +37,6 @@ pub use tool::{ validate_tool_manifest, }; pub use yaml::{ - assert_yaml_parity_subset, assert_yaml_scalar_subset, parse_yaml_document, - yaml_scalar_subset_allows, + assert_execution_profile_yaml_subset, assert_yaml_parity_subset, assert_yaml_scalar_subset, + parse_yaml_document, yaml_scalar_subset_allows, }; diff --git a/crates/runx-parser/src/runner.rs b/crates/runx-parser/src/runner.rs index 0408c65b9..88fcc7651 100644 --- a/crates/runx-parser/src/runner.rs +++ b/crates/runx-parser/src/runner.rs @@ -8,11 +8,14 @@ use crate::skill::{ validate_harness_manifest, validate_runner_definition, }; use crate::{ - ParseError, ValidationError, assert_yaml_parity_subset, + ParseError, ValidationError, assert_execution_profile_yaml_subset, json_fields::{self, JsonFieldReader}, }; const FIELDS: JsonFieldReader = JsonFieldReader::new("runner_manifest"); +const MANIFEST_FIELDS: &[&str] = &[ + "skill", "version", "runx", "policy", "emits", "catalog", "runners", "harness", +]; #[derive(Clone, Debug, PartialEq, Serialize, Deserialize)] pub struct RawRunnerManifestIr { @@ -25,6 +28,14 @@ pub struct SkillRunnerManifest { #[serde(skip_serializing_if = "Option::is_none")] pub skill: Option, #[serde(skip_serializing_if = "Option::is_none")] + pub version: Option, + #[serde(skip_serializing_if = "Option::is_none")] + pub runx: Option, + #[serde(skip_serializing_if = "Option::is_none")] + pub policy: Option, + #[serde(skip_serializing_if = "Option::is_none")] + pub emits: Option, + #[serde(skip_serializing_if = "Option::is_none")] pub catalog: Option, pub runners: BTreeMap, #[serde(skip_serializing_if = "Option::is_none")] @@ -33,7 +44,7 @@ pub struct SkillRunnerManifest { } pub fn parse_runner_manifest_yaml(yaml: &str) -> Result { - assert_yaml_parity_subset("runner_manifest", yaml)?; + assert_execution_profile_yaml_subset("runner_manifest", yaml)?; let parsed: JsonValue = serde_norway::from_str(yaml).map_err(|error| ParseError::InvalidYaml { field: "runner_manifest".to_owned(), @@ -54,6 +65,7 @@ pub fn parse_runner_manifest_yaml(yaml: &str) -> Result Result { + FIELDS.reject_unknown_fields(&raw.document, "runner_manifest", MANIFEST_FIELDS)?; let runners_record = FIELDS.required_object(raw.document.get("runners"), "runners")?; let mut runners = BTreeMap::new(); for (name, value) in runners_record { @@ -74,6 +86,10 @@ pub fn validate_runner_manifest( Ok(SkillRunnerManifest { skill: FIELDS.optional_string(raw.document.get("skill"), "skill")?, + version: FIELDS.optional_string(raw.document.get("version"), "version")?, + runx: FIELDS.optional_object(raw.document.get("runx"), "runx")?, + policy: raw.document.get("policy").cloned(), + emits: raw.document.get("emits").cloned(), catalog: validate_catalog_metadata( FIELDS.optional_object(raw.document.get("catalog"), "catalog")?, "catalog", diff --git a/crates/runx-parser/src/skill.rs b/crates/runx-parser/src/skill.rs index 11368ff81..62748d369 100644 --- a/crates/runx-parser/src/skill.rs +++ b/crates/runx-parser/src/skill.rs @@ -40,6 +40,7 @@ use governance::{ use sandbox::validate_sandbox; use source::default_agent_source; use source::validate_source; +use source::validate_source_fields; const FIELDS: JsonFieldReader = JsonFieldReader::new("skill"); @@ -72,10 +73,14 @@ pub fn validate_skill_with_options( .unwrap_or_else(default_agent_source); let risk = raw.frontmatter.get("risk").cloned(); let governance = validate_skill_governance(&raw, runx.as_ref(), risk.as_ref())?; + let category = validate_portable_skill_category(&raw)?; + let runx_category = validate_runx_skill_category(runx.as_ref())?; Ok(ValidatedSkill { name: FIELDS.required_string(raw.frontmatter.get("name"), "name")?, description: FIELDS.optional_string(raw.frontmatter.get("description"), "description")?, + category, + runx_category, body: raw.body.clone(), source: validate_source(&source, runx.as_ref())?, inputs: validate_inputs( @@ -98,6 +103,28 @@ pub fn validate_skill_with_options( }) } +fn validate_portable_skill_category(raw: &RawSkillIr) -> Result, ValidationError> { + Ok(normalize_optional_category(FIELDS.optional_string( + raw.frontmatter.get("category"), + "category", + )?)) +} + +fn validate_runx_skill_category( + runx: Option<&JsonObject>, +) -> Result, ValidationError> { + Ok(normalize_optional_category(FIELDS.optional_string( + field_value(runx, "category"), + "runx.category", + )?)) +} + +fn normalize_optional_category(value: Option) -> Option { + value + .map(|value| value.trim().to_owned()) + .filter(|value| !value.is_empty()) +} + fn validate_runx_metadata( value: Option<&JsonValue>, mode: ValidateSkillMode, diff --git a/crates/runx-parser/src/skill/catalog.rs b/crates/runx-parser/src/skill/catalog.rs index 5c31d8fe7..06e85090c 100644 --- a/crates/runx-parser/src/skill/catalog.rs +++ b/crates/runx-parser/src/skill/catalog.rs @@ -18,6 +18,7 @@ pub enum CatalogAudience { Public, Builder, Operator, + System, } #[derive(Clone, Copy, Debug, PartialEq, Eq, Serialize, Deserialize)] @@ -53,6 +54,7 @@ impl CatalogAudience { CatalogAudience::Public => "public", CatalogAudience::Builder => "builder", CatalogAudience::Operator => "operator", + CatalogAudience::System => "system", } } } @@ -153,8 +155,9 @@ fn parse_catalog_audience( "public" => Ok(CatalogAudience::Public), "builder" => Ok(CatalogAudience::Builder), "operator" => Ok(CatalogAudience::Operator), + "system" => Ok(CatalogAudience::System), _ => Err(FIELDS.validation_error(format!( - "{label}.audience must be public, builder, or operator." + "{label}.audience must be public, builder, operator, or system." ))), } } diff --git a/crates/runx-parser/src/skill/runner_definition.rs b/crates/runx-parser/src/skill/runner_definition.rs index 516975525..3bbbb3e10 100644 --- a/crates/runx-parser/src/skill/runner_definition.rs +++ b/crates/runx-parser/src/skill/runner_definition.rs @@ -6,17 +6,75 @@ use super::{ FIELDS, SkillGovernance, SkillRunnerDefinition, field_value, first_value, nested_value, validate_allowed_tools, validate_artifact_contract, validate_execution_semantics, validate_idempotency, validate_inputs, validate_mutating, validate_retry, validate_source, + validate_source_fields, }; +const RUNNER_FIELDS: &[&str] = &[ + "act", + "agent", + "agent_card_url", + "agent_identity", + "allowed_tools", + "args", + "arguments", + "artifacts", + "auth", + "catalog_ref", + "command", + "context", + "context_skills", + "cwd", + "default", + "execution", + "external_adapter", + "external_adapter_manifest", + "external_adapter_manifest_path", + "graph", + "headers", + "hook", + "http", + "idempotency", + "input_mode", + "inputs", + "instructions", + "invocation_id", + "method", + "mutating", + "outputs", + "policy", + "retry", + "risk", + "run_id", + "runx", + "runtime", + "sandbox", + "server", + "skill_ref", + "scopes", + "source", + "task", + "timeout_seconds", + "tool", + "type", + "url", + "allow_private_network", +]; + pub(crate) fn validate_runner_definition( name: &str, runner: JsonObject, ) -> Result { + FIELDS.reject_unknown_fields(&runner, &format!("runners.{name}"), RUNNER_FIELDS)?; let runx = FIELDS.optional_object(runner.get("runx"), &format!("runners.{name}.runx"))?; crate::runner::resolve_post_run_reflect_policy(runx.as_ref(), &format!("runners.{name}.runx"))?; - let source_record = FIELDS - .optional_object(runner.get("source"), &format!("runners.{name}.source"))? - .unwrap_or_else(|| runner.clone()); + let source_record = + match FIELDS.optional_object(runner.get("source"), &format!("runners.{name}.source"))? { + Some(source) => { + validate_source_fields(&source, &format!("runners.{name}.source"))?; + source + } + None => runner.clone(), + }; let risk = runner.get("risk").cloned(); let governance = validate_runner_governance(name, &runner, runx.as_ref(), risk.as_ref())?; Ok(SkillRunnerDefinition { diff --git a/crates/runx-parser/src/skill/source.rs b/crates/runx-parser/src/skill/source.rs index ca4148784..d03d5d39e 100644 --- a/crates/runx-parser/src/skill/source.rs +++ b/crates/runx-parser/src/skill/source.rs @@ -8,6 +8,39 @@ use super::{ field_value, first_value, validate_sandbox, }; +const SOURCE_FIELDS: &[&str] = &[ + "act", + "agent", + "agent_card_url", + "agent_identity", + "allow_private_network", + "args", + "arguments", + "catalog_ref", + "command", + "cwd", + "external_adapter", + "external_adapter_manifest", + "external_adapter_manifest_path", + "graph", + "headers", + "hook", + "http", + "input_mode", + "invocation_id", + "method", + "outputs", + "run_id", + "sandbox", + "server", + "skill_ref", + "task", + "timeout_seconds", + "tool", + "type", + "url", +]; + pub fn validate_skill_source( source: &JsonObject, runx: Option<&JsonObject>, @@ -15,6 +48,13 @@ pub fn validate_skill_source( validate_source(source, runx) } +pub(super) fn validate_source_fields( + source: &JsonObject, + field: &str, +) -> Result<(), ValidationError> { + FIELDS.reject_unknown_fields(source, field, SOURCE_FIELDS) +} + pub(super) fn validate_source( source: &JsonObject, runx: Option<&JsonObject>, @@ -225,8 +265,11 @@ fn validate_http_source( if source_type != "http" { return Ok(None); } - let url = FIELDS.required_string(source.get("url"), "source.url")?; - let method = match FIELDS.optional_string(source.get("method"), "source.method")? { + let http = FIELDS + .optional_object(source.get("http"), "source.http")? + .unwrap_or_else(|| source.clone()); + let url = FIELDS.required_string(http.get("url"), "source.url")?; + let method = match FIELDS.optional_string(http.get("method"), "source.method")? { Some(method) => { if !matches!( method.to_ascii_uppercase().as_str(), @@ -243,9 +286,9 @@ fn validate_http_source( Ok(Some(SkillHttpSource { url, method, - headers: validate_http_headers(source.get("headers"))?, + headers: validate_http_headers(http.get("headers"))?, allow_private_network: FIELDS.optional_bool( - source.get("allow_private_network"), + http.get("allow_private_network"), "source.allow_private_network", )?, })) diff --git a/crates/runx-parser/src/skill/types.rs b/crates/runx-parser/src/skill/types.rs index eabe1bb8b..723153339 100644 --- a/crates/runx-parser/src/skill/types.rs +++ b/crates/runx-parser/src/skill/types.rs @@ -158,6 +158,7 @@ pub struct SkillSource { /// the act's structure from these and the trusted inputs; the model authors only /// the reason prose. Absent an `act:` block, a run seals a generic observation. #[derive(Clone, Debug, PartialEq, Eq, Serialize, Deserialize)] +#[serde(deny_unknown_fields)] pub struct ActDeclaration { #[serde(skip_serializing_if = "Option::is_none")] pub form: Option, @@ -180,10 +181,16 @@ pub struct ActDeclaration { #[serde(skip_serializing_if = "Option::is_none")] pub effect_from: Option, #[serde(skip_serializing_if = "Option::is_none")] + pub effect_field_from: Option, + #[serde(skip_serializing_if = "Option::is_none")] + pub effect_from_input: Option, + #[serde(skip_serializing_if = "Option::is_none")] pub effect_type: Option, #[serde(skip_serializing_if = "Option::is_none")] pub effect_prefix: Option, #[serde(skip_serializing_if = "Option::is_none")] + pub effect_prefix_from: Option, + #[serde(skip_serializing_if = "Option::is_none")] pub actor_from: Option, #[serde(skip_serializing_if = "Option::is_none")] pub authority_from: Option, @@ -300,6 +307,10 @@ pub struct ValidatedSkill { pub name: String, #[serde(skip_serializing_if = "Option::is_none")] pub description: Option, + #[serde(skip_serializing_if = "Option::is_none")] + pub category: Option, + #[serde(skip_serializing_if = "Option::is_none")] + pub runx_category: Option, pub body: String, pub source: SkillSource, pub inputs: BTreeMap, diff --git a/crates/runx-parser/src/yaml.rs b/crates/runx-parser/src/yaml.rs index 04f1a3906..a1b673d62 100644 --- a/crates/runx-parser/src/yaml.rs +++ b/crates/runx-parser/src/yaml.rs @@ -1,6 +1,8 @@ // rust-style-allow: large-file the QuoteScanner state machine and its // quote-aware scanners belong next to the parity-subset rules they enforce; // splitting the scanner from the rules trades clarity for two-file traversal. +use std::collections::HashSet; + use serde::de::DeserializeOwned; use crate::ParseError; @@ -20,18 +22,53 @@ where } pub fn assert_yaml_parity_subset(field: &str, source: &str) -> Result<(), ParseError> { + let mut block_scalar_indent = None; for (line_index, line) in source.lines().enumerate() { let line_number = line_index + 1; let Some(content) = strip_yaml_comment(line) else { continue; }; let trimmed = content.trim(); + if let Some(indent) = block_scalar_indent { + if trimmed.is_empty() || leading_spaces(content) > indent { + continue; + } + block_scalar_indent = None; + } if trimmed.is_empty() || trimmed.starts_with("---") || trimmed.starts_with("...") { continue; } reject_explicit_mapping_key(field, line_number, trimmed)?; reject_embedded_colon_key(field, line_number, trimmed)?; reject_colon_space_plain_scalar(field, line_number, content)?; + block_scalar_indent = block_scalar_indent_after(content).or(block_scalar_indent); + } + Ok(()) +} + +pub fn assert_execution_profile_yaml_subset(field: &str, source: &str) -> Result<(), ParseError> { + assert_yaml_parity_subset(field, source)?; + let mut mapping_stack = Vec::new(); + let mut block_scalar_indent = None; + for (line_index, line) in source.lines().enumerate() { + let line_number = line_index + 1; + let Some(content) = strip_yaml_comment(line) else { + continue; + }; + let trimmed = content.trim(); + if let Some(indent) = block_scalar_indent { + if trimmed.is_empty() || leading_spaces(content) > indent { + continue; + } + block_scalar_indent = None; + } + if trimmed.is_empty() { + continue; + } + reject_document_marker(field, line_number, trimmed)?; + reject_yaml_reference_syntax(field, line_number, content)?; + reject_duplicate_mapping_key(field, line_number, content, &mut mapping_stack)?; + block_scalar_indent = block_scalar_indent_after(content).or(block_scalar_indent); } Ok(()) } @@ -222,6 +259,184 @@ fn reject_colon_space_plain_scalar( Ok(()) } +fn reject_document_marker( + field: &str, + line_number: usize, + trimmed: &str, +) -> Result<(), ParseError> { + if trimmed == "---" + || trimmed == "..." + || trimmed.starts_with("--- ") + || trimmed.starts_with("... ") + { + return Err(ParseError::InvalidYaml { + field: field.to_owned(), + message: format!( + "YAML document markers are not supported in X.yaml at line {line_number}; use one plain profile document." + ), + }); + } + Ok(()) +} + +fn reject_yaml_reference_syntax( + field: &str, + line_number: usize, + content: &str, +) -> Result<(), ParseError> { + for token in [": &", ": *", ": !", "- &", "- *", "- !"] { + if contains_plain_token(content, token) { + return Err(ParseError::InvalidYaml { + field: field.to_owned(), + message: format!( + "YAML anchors, aliases, and tags are not supported in X.yaml at line {line_number}; write the profile explicitly." + ), + }); + } + } + let trimmed = content.trim_start(); + if trimmed.starts_with(['&', '*', '!']) { + return Err(ParseError::InvalidYaml { + field: field.to_owned(), + message: format!( + "YAML anchors, aliases, and tags are not supported in X.yaml at line {line_number}; write the profile explicitly." + ), + }); + } + Ok(()) +} + +fn contains_plain_token(content: &str, token: &str) -> bool { + let mut scanner = QuoteScanner::new(); + for (index, char) in content.char_indices() { + if scanner.is_plain_at(char) && content[index..].starts_with(token) { + return true; + } + scanner.consume(char); + } + false +} + +struct MappingFrame { + indent: usize, + keys: HashSet, +} + +fn reject_duplicate_mapping_key( + field: &str, + line_number: usize, + content: &str, + stack: &mut Vec, +) -> Result<(), ParseError> { + let indent = leading_spaces(content); + let trimmed = content.trim_start(); + let (key_indent, key, sequence_item) = match sequence_item_key(trimmed, indent) { + Some(value) => value, + None => { + let Some((key, _)) = top_level_plain_key(trimmed) else { + return Ok(()); + }; + (indent, key, false) + } + }; + if key == "<<" { + return Err(ParseError::InvalidYaml { + field: field.to_owned(), + message: format!( + "YAML merge keys are not supported in X.yaml at line {line_number}; write the profile explicitly." + ), + }); + } + if sequence_item { + while stack.last().is_some_and(|frame| frame.indent >= key_indent) { + stack.pop(); + } + } else { + while stack.last().is_some_and(|frame| frame.indent > key_indent) { + stack.pop(); + } + } + if stack.last().is_none_or(|frame| frame.indent != key_indent) { + stack.push(MappingFrame { + indent: key_indent, + keys: HashSet::new(), + }); + } + let Some(frame) = stack.last_mut() else { + return Err(ParseError::InvalidYaml { + field: field.to_owned(), + message: format!("could not track mapping key {key:?} in X.yaml at line {line_number}"), + }); + }; + if !frame.keys.insert(key.to_owned()) { + return Err(ParseError::InvalidYaml { + field: field.to_owned(), + message: format!( + "duplicate mapping key {key:?} in X.yaml at line {line_number}; keep profile keys unique." + ), + }); + } + Ok(()) +} + +fn sequence_item_key(trimmed: &str, indent: usize) -> Option<(usize, &str, bool)> { + let rest = trimmed.strip_prefix("- ")?; + let item = rest.trim_start(); + let leading = rest.len() - item.len(); + let (key, _) = top_level_plain_key(item)?; + Some((indent + 2 + leading, key, true)) +} + +fn leading_spaces(content: &str) -> usize { + content.bytes().take_while(|byte| *byte == b' ').count() +} + +fn block_scalar_indent_after(content: &str) -> Option { + block_scalar_value_candidates(content) + .iter() + .any(|value| is_block_scalar_header(value)) + .then(|| leading_spaces(content)) +} + +fn block_scalar_value_candidates(content: &str) -> Vec<&str> { + let mut candidates = Vec::new(); + if let Some((_, value)) = split_plain_mapping_value(content) { + candidates.push(value); + } + let trimmed = content.trim_start(); + if let Some(rest) = trimmed.strip_prefix("- ") { + let item = rest.trim_start(); + candidates.push(item); + if let Some((_, value)) = split_plain_mapping_value(item) { + candidates.push(value); + } + } + candidates +} + +fn is_block_scalar_header(value: &str) -> bool { + let trimmed = value.trim(); + let mut chars = trimmed.chars(); + let Some(first) = chars.next() else { + return false; + }; + if !matches!(first, '|' | '>') { + return false; + } + let mut seen_chomp = false; + let mut seen_indent = false; + for char in chars { + if matches!(char, '+' | '-') && !seen_chomp { + seen_chomp = true; + } else if char.is_ascii_digit() && !seen_indent { + seen_indent = true; + } else { + return false; + } + } + true +} + fn split_plain_mapping_value(content: &str) -> Option<(&str, &str)> { let trimmed = content.trim_start(); let (key, delimiter_index) = top_level_plain_key(trimmed)?; @@ -303,7 +518,10 @@ fn is_special_float(value: &str) -> bool { #[cfg(test)] mod tests { - use super::{assert_yaml_parity_subset, assert_yaml_scalar_subset, yaml_scalar_subset_allows}; + use super::{ + assert_execution_profile_yaml_subset, assert_yaml_parity_subset, assert_yaml_scalar_subset, + yaml_scalar_subset_allows, + }; #[test] fn scalar_subset_rejects_divergent_forms() { @@ -359,6 +577,50 @@ mod tests { assert!(result.is_err(), "expected rejection, got {result:?}"); } + #[test] + fn execution_profile_subset_rejects_yaml_references_and_document_markers() { + for literal in [ + "---\nskill: example", + "runners:\n one:\n outputs: &shared\n result: string", + "runners:\n one:\n outputs: *shared", + "runners:\n one:\n runx:\n <<: *shared", + "runners:\n one:\n type: !custom graph", + ] { + let result = assert_execution_profile_yaml_subset("runner_manifest", literal); + assert!(result.is_err(), "expected rejection, got {result:?}"); + } + } + + #[test] + fn execution_profile_subset_rejects_duplicate_keys_but_allows_sequence_reuse() { + let result = + assert_execution_profile_yaml_subset("runner_manifest", "skill: one\nskill: two\n"); + assert!( + result.is_err(), + "expected duplicate key rejection, got {result:?}" + ); + + let sequence_result = assert_execution_profile_yaml_subset( + "runner_manifest", + r#" +runners: + demo: + type: graph + graph: + name: demo + steps: + - id: first + tool: one.tool + - id: second + tool: two.tool +"#, + ); + assert!( + sequence_result.is_ok(), + "sequence item maps may reuse keys in separate items: {sequence_result:?}" + ); + } + // Regression cases for the single-quote `''` escape. The earlier toggle // flipped on every `'`, so `'it''s'` mis-segmented into three scalars and // any `:` after byte 4 was treated as still-quoted. diff --git a/crates/runx-parser/tests/parser_catalog.rs b/crates/runx-parser/tests/parser_catalog.rs index 7cb17b6d6..c8d990247 100644 --- a/crates/runx-parser/tests/parser_catalog.rs +++ b/crates/runx-parser/tests/parser_catalog.rs @@ -69,6 +69,30 @@ runners: Ok(()) } +#[test] +fn catalog_audience_accepts_system_sync_profiles() -> Result<(), String> { + let manifest = parse_manifest( + r#" +catalog: + kind: skill + audience: system + visibility: internal + role: canonical +runners: + default: + source: + type: agent +"#, + )?; + + let catalog = manifest + .catalog + .ok_or_else(|| "expected catalog metadata".to_owned())?; + assert_eq!(catalog.audience, CatalogAudience::System); + assert_eq!(catalog.audience.as_str(), "system"); + Ok(()) +} + #[test] fn unknown_catalog_kind_fails_closed() -> Result<(), String> { let raw = parse_runner_manifest_yaml( diff --git a/crates/runx-parser/tests/parser_fixtures.rs b/crates/runx-parser/tests/parser_fixtures.rs index 7d50a3dc4..fe8c8f05e 100644 --- a/crates/runx-parser/tests/parser_fixtures.rs +++ b/crates/runx-parser/tests/parser_fixtures.rs @@ -163,6 +163,30 @@ fn skill_fixtures_match_typescript() -> Result<(), String> { Ok(()) } +#[test] +fn skill_category_accepts_portable_top_level_field() -> Result<(), String> { + let skill = validate_skill(parse_skill_markdown( + "---\nname: docs-demo\ndescription: Demo skill.\ncategory: documentation\n---\n# Demo\n", + ).map_err(|error| error.to_string())?) + .map_err(|error| error.to_string())?; + + assert_eq!(skill.category.as_deref(), Some("documentation")); + assert_eq!(skill.runx_category.as_deref(), None); + Ok(()) +} + +#[test] +fn skill_category_accepts_runx_catalog_override() -> Result<(), String> { + let skill = validate_skill(parse_skill_markdown( + "---\nname: docs-demo\ndescription: Demo skill.\ncategory: documentation\nrunx:\n category: content\n---\n# Demo\n", + ).map_err(|error| error.to_string())?) + .map_err(|error| error.to_string())?; + + assert_eq!(skill.category.as_deref(), Some("documentation")); + assert_eq!(skill.runx_category.as_deref(), Some("content")); + Ok(()) +} + #[test] fn runner_manifest_fixtures_match_typescript() -> Result<(), String> { for fixture_json in RUNNER_MANIFEST_FIXTURES { diff --git a/crates/runx-runtime/Cargo.toml b/crates/runx-runtime/Cargo.toml index e609d49be..5c6c8fdd9 100644 --- a/crates/runx-runtime/Cargo.toml +++ b/crates/runx-runtime/Cargo.toml @@ -53,7 +53,11 @@ runx-contracts.workspace = true runx-core.workspace = true runx-parser.workspace = true runx-receipts.workspace = true -reqwest = { version = "=0.13.3", default-features = false, features = ["rustls-no-provider", "json"], optional = true } +# gzip/brotli/deflate/zstd + http2 so the governed fetch transport negotiates and +# decodes like a real browser; a no-compression, http1-only client is a bot tell to +# Cloudflare and friends. The response cap measures DECODED bytes (streaming guard in +# read_limited_response_body), so a decompression bomb stays bounded. +reqwest = { version = "=0.13.3", default-features = false, features = ["rustls-no-provider", "json", "gzip", "brotli", "deflate", "zstd", "http2"], optional = true } ring = "0.17.14" # Drive rustls with the ring provider (already linked via `ring`) instead of # reqwest's default aws-lc-rs, so the vendored aws-lc-sys C crypto blob is not diff --git a/crates/runx-runtime/src/adapters/agent_resolver.rs b/crates/runx-runtime/src/adapters/agent_resolver.rs index 717ebbe04..d193e175a 100644 --- a/crates/runx-runtime/src/adapters/agent_resolver.rs +++ b/crates/runx-runtime/src/adapters/agent_resolver.rs @@ -9,7 +9,9 @@ use std::collections::BTreeMap; use std::path::PathBuf; -use runx_contracts::{ContextEntry, JsonObject, JsonValue, ResolutionRequest}; +use runx_contracts::{ + ContextEntry, JsonObject, JsonValue, OutputField, OutputType, ResolutionRequest, +}; use super::agent::{AgentResolution, AgentResolver, AgentResolverError}; use super::agent_anthropic::{AgentToolDefinition, AnthropicModelCaller}; @@ -62,10 +64,92 @@ fn object_schema() -> JsonValue { JsonValue::Object(schema) } +fn final_result_schema(output: Option<&BTreeMap>) -> JsonValue { + let Some(output) = output else { + return object_schema(); + }; + if output.is_empty() { + return object_schema(); + } + + let mut properties = JsonObject::new(); + let mut required = Vec::new(); + for (name, field) in output { + properties.insert(name.clone(), output_field_schema(field)); + if output_field_required(field) { + required.push(JsonValue::String(name.clone())); + } + } + + let mut schema = JsonObject::new(); + schema.insert("type".to_owned(), JsonValue::String("object".to_owned())); + schema.insert("properties".to_owned(), JsonValue::Object(properties)); + schema.insert("additionalProperties".to_owned(), JsonValue::Bool(false)); + if !required.is_empty() { + schema.insert("required".to_owned(), JsonValue::Array(required)); + } + JsonValue::Object(schema) +} + +fn output_field_required(field: &OutputField) -> bool { + match field { + OutputField::Type(_) => true, + OutputField::Spec(spec) => spec.required.unwrap_or(true), + } +} + +fn output_field_schema(field: &OutputField) -> JsonValue { + let mut schema = JsonObject::new(); + match field { + OutputField::Type(field_type) => { + schema.insert( + "type".to_owned(), + JsonValue::String(output_type_name(field_type).to_owned()), + ); + } + OutputField::Spec(spec) => { + if let Some(field_type) = spec.field_type.as_ref() { + schema.insert( + "type".to_owned(), + JsonValue::String(output_type_name(field_type).to_owned()), + ); + } + if let Some(values) = spec.enum_values.as_ref() { + schema.insert( + "enum".to_owned(), + JsonValue::Array(values.iter().cloned().map(JsonValue::String).collect()), + ); + } + if let Some(description) = spec.description.as_ref() { + schema.insert( + "description".to_owned(), + JsonValue::String(description.clone()), + ); + } + } + } + JsonValue::Object(schema) +} + +const fn output_type_name(field_type: &OutputType) -> &'static str { + match field_type { + OutputType::String => "string", + OutputType::Number => "number", + OutputType::Integer => "integer", + OutputType::Boolean => "boolean", + OutputType::Array => "array", + OutputType::Object => "object", + OutputType::Null => "null", + } +} + /// The skill's allowed tools plus the final-result tool the model calls to finish. /// Input schemas are permissive for now; resolving each tool's manifest schema is /// a refinement, not required for the loop to run governed. -fn tool_definitions<'a>(tool_names: impl Iterator) -> Vec { +fn tool_definitions<'a>( + tool_names: impl Iterator, + output: Option<&BTreeMap>, +) -> Vec { let mut tools: Vec = tool_names .map(|name| AgentToolDefinition { name: name.to_owned(), @@ -76,7 +160,7 @@ fn tool_definitions<'a>(tool_names: impl Iterator) -> Vec AgentResolver for AnthropicAgentResolver = tools.iter().map(|tool| tool.name.as_str()).collect(); assert!( names == ["pay", "read", FINAL_RESULT_TOOL], @@ -181,6 +268,44 @@ mod tests { ); } + #[test] + fn final_result_schema_uses_declared_outputs() -> Result<(), String> { + let outputs = BTreeMap::from([ + ("decision".to_owned(), OutputField::Type(OutputType::String)), + ("quality".to_owned(), OutputField::Type(OutputType::Object)), + ]); + let tools = tool_definitions([].into_iter(), Some(&outputs)); + let final_tool = tools + .iter() + .find(|tool| tool.name == FINAL_RESULT_TOOL) + .ok_or_else(|| "missing final-result tool".to_owned())?; + + let JsonValue::Object(schema) = &final_tool.input_schema else { + return Err("final result schema should be an object".to_owned()); + }; + assert_eq!( + schema.get("type"), + Some(&JsonValue::String("object".to_owned())) + ); + let Some(JsonValue::Object(properties)) = schema.get("properties") else { + return Err("properties should be an object".to_owned()); + }; + assert!(properties.contains_key("decision")); + assert!(properties.contains_key("quality")); + assert_eq!( + schema.get("required"), + Some(&JsonValue::Array(vec![ + JsonValue::String("decision".to_owned()), + JsonValue::String("quality".to_owned()), + ])) + ); + assert_eq!( + schema.get("additionalProperties"), + Some(&JsonValue::Bool(false)) + ); + Ok(()) + } + #[test] fn prompt_carries_instructions_directive_and_inputs() { let mut inputs = JsonObject::new(); diff --git a/crates/runx-runtime/src/adapters/http.rs b/crates/runx-runtime/src/adapters/http.rs index 97e095aaf..69fd33436 100644 --- a/crates/runx-runtime/src/adapters/http.rs +++ b/crates/runx-runtime/src/adapters/http.rs @@ -26,12 +26,18 @@ use crate::adapter::{ }; use crate::credentials::SecretEnv; use crate::runtime_http::{ - HttpMethod, ReqwestHttpTransport, RuntimeHttpHeader, RuntimeHttpRequest, RuntimeHttpTransport, + DEFAULT_BROWSER_USER_AGENT, HttpMethod, ReqwestHttpTransport, RuntimeHttpHeader, + RuntimeHttpRequest, RuntimeHttpTransport, }; use runx_parser::SourceKind; const HTTP_SKILL: &str = "http"; const RUNX_HTTP_ALLOW_PRIVATE_NETWORK_ENV: &str = "RUNX_HTTP_ALLOW_PRIVATE_NETWORK"; +// The open-web fetch surface presents a browser profile by default so a Cloudflare-fronted +// site does not score us as a bot. RUNX_HTTP_BROWSER=0 opts a run out (back to the plain +// client); RUNX_HTTP_USER_AGENT overrides the UA string. +const RUNX_HTTP_BROWSER_ENV: &str = "RUNX_HTTP_BROWSER"; +const RUNX_HTTP_USER_AGENT_ENV: &str = "RUNX_HTTP_USER_AGENT"; /// A governed HTTP call: a method, a URL, and the request headers (auth and the /// like, already resolved). Inputs are mapped to the query string (GET, DELETE) or @@ -123,8 +129,9 @@ fn resolve_path_template( Ok((out, remaining)) } -fn json_body(inputs: &JsonObject) -> Result { - serde_json::to_string(&serde_json::to_value(inputs).unwrap_or(WireValue::Null)) +fn json_body(inputs: &JsonObject, secrets: &SecretEnv) -> Result { + let value = substitute_json_secrets(&JsonValue::Object(inputs.clone()), secrets)?; + serde_json::to_string(&serde_json::to_value(value).unwrap_or(WireValue::Null)) .map_err(|error| failure(format!("serializing http request body: {error}"))) } @@ -134,6 +141,7 @@ pub fn execute_http_call( transport: &T, call: &HttpCall, inputs: &JsonObject, + secrets: &SecretEnv, ) -> Result { let (resolved_url, query_inputs) = resolve_path_template(&call.url, inputs)?; let mut headers = call.headers.clone(); @@ -145,7 +153,7 @@ pub fn execute_http_call( { headers.push(RuntimeHttpHeader::new("content-type", "application/json")); } - (resolved_url, Some(json_body(&query_inputs)?)) + (resolved_url, Some(json_body(&query_inputs, secrets)?)) } HttpMethod::Get | HttpMethod::Delete => (with_query(&resolved_url, &query_inputs), None), }; @@ -179,22 +187,22 @@ pub fn execute_http_call( const SECRET_PREFIX: &str = "${secret:"; -/// Resolve `${secret:NAME}` references in a header value against the run's secret -/// env, mirroring how the cli-tool front lets a command reference a delivered -/// secret. A reference to a secret that was not delivered fails closed. +/// Resolve `${secret:NAME}` references against the run's secret env, mirroring +/// how the cli-tool front lets a command reference a delivered secret. A +/// reference to a secret that was not delivered fails closed. fn substitute_secrets(value: &str, secrets: &SecretEnv) -> Result { let mut out = String::with_capacity(value.len()); let mut rest = value; while let Some(start) = rest.find(SECRET_PREFIX) { out.push_str(&rest[..start]); let after = &rest[start + SECRET_PREFIX.len()..]; - let end = after.find('}').ok_or_else(|| { - failure("http header secret reference is missing a closing '}'".to_owned()) - })?; + let end = after + .find('}') + .ok_or_else(|| failure("http secret reference is missing a closing '}'".to_owned()))?; let name = &after[..end]; let secret = secrets.get(name).ok_or_else(|| { failure(format!( - "http header references secret {name}, which was not delivered to this run" + "http references secret {name}, which was not delivered to this run" )) })?; out.push_str(secret); @@ -204,6 +212,26 @@ fn substitute_secrets(value: &str, secrets: &SecretEnv) -> Result Result { + match value { + JsonValue::String(value) => substitute_secrets(value, secrets).map(JsonValue::String), + JsonValue::Array(values) => values + .iter() + .map(|value| substitute_json_secrets(value, secrets)) + .collect::, _>>() + .map(JsonValue::Array), + JsonValue::Object(object) => object + .iter() + .map(|(key, value)| Ok((key.clone(), substitute_json_secrets(value, secrets)?))) + .collect::>() + .map(JsonValue::Object), + value => Ok(value.clone()), + } +} + /// Build the request headers from the source's validated `headers` map, resolving /// any `${secret:NAME}` references. Header names and values are otherwise passed /// through verbatim; the transport validates them and redacts sensitive ones. @@ -255,6 +283,25 @@ fn operator_allows_private_network(env: &BTreeMap) -> bool { .is_some_and(|value| matches!(value.as_str(), "1" | "true" | "yes" | "on")) } +/// The browser User-Agent for the open-web fetch surface, or `None` (the plain client) +/// when the run opts out with `RUNX_HTTP_BROWSER=0`. `RUNX_HTTP_USER_AGENT` overrides the +/// default Chrome string. Browser-on is the default. +fn browser_user_agent(env: &BTreeMap) -> Option { + let opted_out = env + .get(RUNX_HTTP_BROWSER_ENV) + .map(|value| value.trim().to_ascii_lowercase()) + .is_some_and(|value| matches!(value.as_str(), "0" | "false" | "no" | "off")); + if opted_out { + return None; + } + let user_agent = env + .get(RUNX_HTTP_USER_AGENT_ENV) + .map(|value| value.trim()) + .filter(|value| !value.is_empty()) + .unwrap_or(DEFAULT_BROWSER_USER_AGENT); + Some(user_agent.to_owned()) +} + /// The governed HTTP skill adapter: reads `url`/`method`/`headers` from an `http` /// source, resolves credential headers, and runs the call through the governed /// transport. The default constructs a [`ReqwestHttpTransport`]; the engine itself @@ -295,13 +342,19 @@ impl SkillAdapter for HttpSkillAdapter { "http source requested private-network access but operator grant {RUNX_HTTP_ALLOW_PRIVATE_NETWORK_ENV}=1 is not set" ))); } - let transport = if allow_private_network { - ReqwestHttpTransport::with_private_network_access() - } else { - ReqwestHttpTransport::new() - } + // The http tool is the open-web fetch surface, so it presents the browser + // profile by default; a per-source header still overrides any browser default. + let transport = ReqwestHttpTransport::with_options( + allow_private_network, + browser_user_agent(&request.env), + ) .map_err(|error| failure(format!("http transport unavailable: {error}")))?; - let mut output = execute_http_call(&transport, &call, &merged_inputs(&request))?; + let mut output = execute_http_call( + &transport, + &call, + &merged_inputs(&request), + request.credential_delivery.secret_env(), + )?; add_credential_delivery_metadata(&mut output, &request.credential_delivery)?; Ok(output) } @@ -370,6 +423,10 @@ mod tests { .collect() } + fn empty_secrets() -> SecretEnv { + SecretEnv::default() + } + fn http_invocation( allow_private_network: Option, env: BTreeMap, @@ -441,7 +498,12 @@ mod tests { url: "https://api.example.test/v1/pets".to_owned(), headers: Vec::new(), }; - let output = execute_http_call(&transport, &call, &inputs(&[("id", "p-7")]))?; + let output = execute_http_call( + &transport, + &call, + &inputs(&[("id", "p-7")]), + &empty_secrets(), + )?; assert_eq!(output.status, InvocationStatus::Success); assert_eq!(output.stdout, r#"{"ok":true}"#); let sent = transport.requests.borrow(); @@ -461,7 +523,12 @@ mod tests { url: "https://api.example.test/v1/pets".to_owned(), headers: Vec::new(), }; - execute_http_call(&transport, &call, &inputs(&[("name", "rex")]))?; + execute_http_call( + &transport, + &call, + &inputs(&[("name", "rex")]), + &empty_secrets(), + )?; let sent = transport.requests.borrow(); assert!( sent[0] @@ -474,6 +541,41 @@ mod tests { Ok(()) } + #[test] + fn post_substitutes_secret_references_in_json_body() -> Result<(), RuntimeError> { + let delivery = crate::credentials::CredentialDelivery::from_local_descriptor( + "api", + "bearer", + "API_TOKEN", + "credential:test", + vec!["api.call".to_owned()], + "api_secret", + ) + .map_err(|error| failure(format!("building test credential: {error}")))?; + let transport = stub(201, ""); + let call = HttpCall { + method: HttpMethod::Post, + url: "https://api.example.test/v1/pets".to_owned(), + headers: Vec::new(), + }; + execute_http_call( + &transport, + &call, + &inputs(&[("token", "${secret:API_TOKEN}")]), + delivery.secret_env(), + )?; + let sent = transport.requests.borrow(); + assert!( + sent[0] + .body + .as_deref() + .is_some_and(|body| body.contains(r#""token":"api_secret""#)), + "POST body should substitute delivered secret refs; got: {:?}", + sent[0].body + ); + Ok(()) + } + #[test] fn path_template_substitutes_inputs_and_drops_them_from_the_query() -> Result<(), RuntimeError> { @@ -487,6 +589,7 @@ mod tests { &transport, &call, &inputs(&[("id", "p-7"), ("fields", "name")]), + &empty_secrets(), )?; let sent = transport.requests.borrow(); assert!( @@ -508,23 +611,47 @@ mod tests { headers: Vec::new(), }; assert!( - execute_http_call(&transport, &call, &JsonObject::new()).is_err(), + execute_http_call(&transport, &call, &JsonObject::new(), &empty_secrets()).is_err(), "a placeholder with no matching input must fail closed" ); assert!( - execute_http_call(&transport, &call, &inputs(&[("id", "a/b")])).is_err(), + execute_http_call( + &transport, + &call, + &inputs(&[("id", "a/b")]), + &empty_secrets() + ) + .is_err(), "a path value with a path separator must fail closed" ); assert!( - execute_http_call(&transport, &call, &inputs(&[("id", "a#b")])).is_err(), + execute_http_call( + &transport, + &call, + &inputs(&[("id", "a#b")]), + &empty_secrets() + ) + .is_err(), "a path value with a fragment delimiter must fail closed" ); assert!( - execute_http_call(&transport, &call, &inputs(&[("id", "a%2Fb")])).is_err(), + execute_http_call( + &transport, + &call, + &inputs(&[("id", "a%2Fb")]), + &empty_secrets() + ) + .is_err(), "a path value with an encoded path delimiter must fail closed" ); assert!( - execute_http_call(&transport, &call, &inputs(&[("id", "a%3Fb")])).is_err(), + execute_http_call( + &transport, + &call, + &inputs(&[("id", "a%3Fb")]), + &empty_secrets() + ) + .is_err(), "a path value with an encoded query delimiter must fail closed" ); } @@ -537,7 +664,12 @@ mod tests { url: "https://api.example.test/v1/pets/p-7".to_owned(), headers: Vec::new(), }; - execute_http_call(&transport, &call, &inputs(&[("name", "rex")]))?; + execute_http_call( + &transport, + &call, + &inputs(&[("name", "rex")]), + &empty_secrets(), + )?; let sent = transport.requests.borrow(); assert!( sent[0].method == HttpMethod::Put @@ -559,7 +691,12 @@ mod tests { url: "https://api.example.test/v1/pets/p-7".to_owned(), headers: Vec::new(), }; - execute_http_call(&transport, &call, &inputs(&[("name", "rex")]))?; + execute_http_call( + &transport, + &call, + &inputs(&[("name", "rex")]), + &empty_secrets(), + )?; let sent = transport.requests.borrow(); assert!( sent[0].method == HttpMethod::Patch @@ -581,7 +718,12 @@ mod tests { url: "https://api.example.test/v1/pets".to_owned(), headers: Vec::new(), }; - execute_http_call(&transport, &call, &inputs(&[("id", "p-7")]))?; + execute_http_call( + &transport, + &call, + &inputs(&[("id", "p-7")]), + &empty_secrets(), + )?; let sent = transport.requests.borrow(); assert!( sent[0].method == HttpMethod::Delete @@ -605,7 +747,12 @@ mod tests { RuntimeHttpHeader::new("content-type", "application/cbor"), ], }; - execute_http_call(&transport, &call, &inputs(&[("name", "rex")]))?; + execute_http_call( + &transport, + &call, + &inputs(&[("name", "rex")]), + &empty_secrets(), + )?; let sent = transport.requests.borrow(); let content_types = sent[0] .headers @@ -691,7 +838,7 @@ mod tests { url: "https://api.example.test/v1/pets/none".to_owned(), headers: Vec::new(), }; - let output = execute_http_call(&transport, &call, &JsonObject::new())?; + let output = execute_http_call(&transport, &call, &JsonObject::new(), &empty_secrets())?; assert_eq!(output.status, InvocationStatus::Failure); assert_eq!(output.stdout, "not found"); Ok(()) @@ -705,7 +852,7 @@ mod tests { url: "https://api.example.test/v1/pets".to_owned(), headers: Vec::new(), }; - let output = execute_http_call(&transport, &call, &JsonObject::new())?; + let output = execute_http_call(&transport, &call, &JsonObject::new(), &empty_secrets())?; assert_eq!( output.status, InvocationStatus::Failure, diff --git a/crates/runx-runtime/src/adapters/mcp/server_skill.rs b/crates/runx-runtime/src/adapters/mcp/server_skill.rs index 2d0d2d29f..8dbebf77e 100644 --- a/crates/runx-runtime/src/adapters/mcp/server_skill.rs +++ b/crates/runx-runtime/src/adapters/mcp/server_skill.rs @@ -68,6 +68,11 @@ pub(super) fn load_mcp_server_tool( ) -> Result { let skill_path = canonical_skill_path(skill_path)?; let skill = load_skill_for_mcp(&skill_path)?; + let required_scopes = + required_scopes_from_skill(&skill).map_err(|error| RuntimeError::SkillFailed { + skill_name: skill.name.clone(), + message: format!("invalid required scopes: {error}"), + })?; Ok(McpServerTool { name: skill.name.clone(), description: skill @@ -75,7 +80,7 @@ pub(super) fn load_mcp_server_tool( .clone() .unwrap_or_else(|| format!("runx skill {}", skill.name)), input_schema: skill_inputs_to_json_schema(&skill.inputs), - required_scopes: required_scopes_from_skill(&skill), + required_scopes, result: McpServerToolBehavior::Skill(Box::new(McpServerSkillExecution { skill_path, skill, diff --git a/crates/runx-runtime/src/agent_invocation.rs b/crates/runx-runtime/src/agent_invocation.rs index c10db9df7..3040afeab 100644 --- a/crates/runx-runtime/src/agent_invocation.rs +++ b/crates/runx-runtime/src/agent_invocation.rs @@ -85,7 +85,7 @@ fn envelope( skill: skill_name(request, source_type).into(), instructions: envelope_instructions(request).into(), inputs: request.inputs.clone(), - allowed_tools: envelope_allowed_tools(request), + allowed_tools: envelope_allowed_tools(request)?, current_context: request.current_context.clone(), historical_context: Vec::new(), provenance: Vec::new(), @@ -116,20 +116,43 @@ fn envelope_instructions(request: &SkillInvocation) -> String { }) } -fn envelope_allowed_tools(request: &SkillInvocation) -> Vec { - request - .source - .raw - .get("allowed_tools") - .and_then(JsonValue::as_array) - .map(|tools| { - tools - .iter() - .filter_map(JsonValue::as_str) - .filter_map(|value| NonEmptyString::new(value.to_owned())) - .collect::>() - }) - .unwrap_or_default() +fn envelope_allowed_tools(request: &SkillInvocation) -> Result, RuntimeError> { + let Some(value) = request.source.raw.get("allowed_tools") else { + return Ok(Vec::new()); + }; + let JsonValue::Array(tools) = value else { + return Err(invalid_agent_invocation( + request, + "allowed_tools must be an array of non-empty strings", + )); + }; + let mut allowed_tools = Vec::new(); + for (index, value) in tools.iter().enumerate() { + let Some(tool) = value + .as_str() + .map(str::trim) + .filter(|value| !value.is_empty()) + .and_then(|value| NonEmptyString::new(value.to_owned())) + else { + return Err(invalid_agent_invocation( + request, + format!("allowed_tools[{index}] must be a non-empty string"), + )); + }; + allowed_tools.push(tool); + } + Ok(allowed_tools) +} + +fn invalid_agent_invocation(request: &SkillInvocation, message: impl Into) -> RuntimeError { + RuntimeError::SkillFailed { + skill_name: if request.skill_name.is_empty() { + "agent".to_owned() + } else { + request.skill_name.clone() + }, + message: message.into(), + } } fn optional_non_empty(value: Option<&str>) -> Option { diff --git a/crates/runx-runtime/src/effects/provider_permission.rs b/crates/runx-runtime/src/effects/provider_permission.rs index 814248ffd..df3aaea3c 100644 --- a/crates/runx-runtime/src/effects/provider_permission.rs +++ b/crates/runx-runtime/src/effects/provider_permission.rs @@ -85,7 +85,7 @@ fn provider_permission_plan( request: &EffectStepRequest<'_>, policy: &JsonObject, ) -> Result, RuntimeEffectError> { - let verb = verb_field(policy).unwrap_or_else(|| default_verb(request.step.mutating)); + let verb = required_verb_field(policy)?; if policy.contains_key("granted_scopes") { return Err(RuntimeEffectError::Denied { family: PROVIDER_PERMISSION_EFFECT_FAMILY.to_owned(), @@ -93,8 +93,7 @@ fn provider_permission_plan( message: "provider_permission.granted_scopes is self-attested by the graph policy; provide granted scopes through the operator grant environment instead".to_owned(), }); } - let required_scopes = string_array_field(policy, "required_scopes") - .filter(|scopes| !scopes.is_empty()) + let required_scopes = string_array_field(policy, "required_scopes")? .unwrap_or_else(|| request.step.scopes.clone()); if required_scopes.is_empty() { return Ok(None); @@ -125,14 +124,6 @@ fn provider_permission_plan( })) } -fn default_verb(mutating: bool) -> AuthorityVerb { - if mutating { - AuthorityVerb::Write - } else { - AuthorityVerb::Read - } -} - fn provider_permission_denial( request: &EffectStepRequest<'_>, plan: &ProviderPermissionPlan, @@ -177,16 +168,32 @@ fn string_field<'a>(object: &'a JsonObject, key: &str) -> Option<&'a str> { object.get(key).and_then(JsonValue::as_str) } -fn string_array_field(object: &JsonObject, key: &str) -> Option> { - Some( - object - .get(key)? - .as_array()? - .iter() - .filter_map(JsonValue::as_str) - .map(str::to_owned) - .collect(), - ) +fn string_array_field( + object: &JsonObject, + key: &str, +) -> Result>, RuntimeEffectError> { + let Some(value) = object.get(key) else { + return Ok(None); + }; + let Some(values) = value.as_array() else { + return Err(provider_permission_policy_error(format!( + "{key} must be an array" + ))); + }; + values + .iter() + .enumerate() + .map(|(index, value)| match value { + JsonValue::String(scope) if !scope.trim().is_empty() => Ok(scope.trim().to_owned()), + JsonValue::String(_) => Err(provider_permission_policy_error(format!( + "{key}[{index}] must be a non-empty string" + ))), + _ => Err(provider_permission_policy_error(format!( + "{key}[{index}] must be a string" + ))), + }) + .collect::, _>>() + .map(Some) } fn provider_grant_id( @@ -221,18 +228,38 @@ fn parse_scope_list(value: &str) -> Vec { .collect() } -fn verb_field(object: &JsonObject) -> Option { - match string_field(object, "verb")? { - "read" => Some(AuthorityVerb::Read), - "write" => Some(AuthorityVerb::Write), - "comment" => Some(AuthorityVerb::Comment), - "review" => Some(AuthorityVerb::Review), - "merge" => Some(AuthorityVerb::Merge), - "create" => Some(AuthorityVerb::Create), - "update" => Some(AuthorityVerb::Update), - "delete" => Some(AuthorityVerb::Delete), - "execute" => Some(AuthorityVerb::Execute), - _ => None, +fn required_verb_field(object: &JsonObject) -> Result { + let Some(value) = object.get("verb") else { + return Err(provider_permission_policy_error( + "verb is required".to_owned(), + )); + }; + let Some(verb) = value.as_str() else { + return Err(provider_permission_policy_error( + "verb must be a string".to_owned(), + )); + }; + match verb { + "read" => Ok(AuthorityVerb::Read), + "write" => Ok(AuthorityVerb::Write), + "comment" => Ok(AuthorityVerb::Comment), + "review" => Ok(AuthorityVerb::Review), + "merge" => Ok(AuthorityVerb::Merge), + "create" => Ok(AuthorityVerb::Create), + "update" => Ok(AuthorityVerb::Update), + "delete" => Ok(AuthorityVerb::Delete), + "execute" => Ok(AuthorityVerb::Execute), + _ => Err(provider_permission_policy_error(format!( + "verb {verb:?} is not supported" + ))), + } +} + +fn provider_permission_policy_error(message: String) -> RuntimeEffectError { + RuntimeEffectError::Failed { + family: PROVIDER_PERMISSION_EFFECT_FAMILY.to_owned(), + operation: "parse provider permission policy", + message, } } @@ -404,6 +431,77 @@ mod tests { } } + #[test] + fn rejects_missing_or_unknown_policy_verb() -> Result<(), io::Error> { + let effect = ProviderPermissionEffect; + let inputs = JsonObject::new(); + let env = provider_env("github-mcp-read", "repo.read"); + + let mut missing_verb = test_step("read_issue", vec!["repo.read"], false, "read", false); + provider_permission_policy_mut(&mut missing_verb)?.remove("verb"); + let error = match effect.admit(EffectStepRequest { + step: &missing_verb, + inputs: &inputs, + env: &env, + graph_dir: Path::new("."), + }) { + Ok(_) => { + return Err(io::Error::other( + "missing provider permission verb should fail", + )); + } + Err(error) => error, + }; + assert_policy_error(error, "verb is required")?; + + let unknown_verb = test_step("read_issue", vec!["repo.read"], false, "publish", false); + let error = match effect.admit(EffectStepRequest { + step: &unknown_verb, + inputs: &inputs, + env: &env, + graph_dir: Path::new("."), + }) { + Ok(_) => { + return Err(io::Error::other( + "unknown provider permission verb should fail", + )); + } + Err(error) => error, + }; + assert_policy_error(error, "not supported") + } + + #[test] + fn rejects_malformed_required_scopes() -> Result<(), io::Error> { + let effect = ProviderPermissionEffect; + let mut step = test_step("read_issue", vec!["repo.read"], false, "read", false); + provider_permission_policy_mut(&mut step)?.insert( + "required_scopes".to_owned(), + JsonValue::Array(vec![ + JsonValue::String("repo.read".to_owned()), + JsonValue::Bool(false), + ]), + ); + let inputs = JsonObject::new(); + let env = provider_env("github-mcp-read", "repo.read"); + + let error = match effect.admit(EffectStepRequest { + step: &step, + inputs: &inputs, + env: &env, + graph_dir: Path::new("."), + }) { + Ok(_) => { + return Err(io::Error::other( + "malformed provider permission required_scopes should fail", + )); + } + Err(error) => error, + }; + + assert_policy_error(error, "required_scopes[1] must be a string") + } + fn test_step( id: &str, required_scopes: Vec<&str>, @@ -478,4 +576,35 @@ mod tests { .into_iter() .collect() } + + fn provider_permission_policy_mut(step: &mut GraphStep) -> Result<&mut JsonObject, io::Error> { + let Some(value) = step + .policy + .as_mut() + .and_then(|policy| policy.get_mut(PROVIDER_PERMISSION_EFFECT_FAMILY)) + else { + return Err(io::Error::other( + "test step should carry provider permission policy", + )); + }; + let JsonValue::Object(object) = value else { + return Err(io::Error::other( + "test step provider permission policy should be an object", + )); + }; + Ok(object) + } + + fn assert_policy_error(error: RuntimeEffectError, needle: &str) -> Result<(), io::Error> { + match error { + RuntimeEffectError::Failed { + family, + operation: "parse provider permission policy", + message, + } if family == PROVIDER_PERMISSION_EFFECT_FAMILY && message.contains(needle) => Ok(()), + other => Err(io::Error::other(format!( + "unexpected provider permission policy error: {other:?}" + ))), + } + } } diff --git a/crates/runx-runtime/src/execution.rs b/crates/runx-runtime/src/execution.rs index 46e56ab75..26a508840 100644 --- a/crates/runx-runtime/src/execution.rs +++ b/crates/runx-runtime/src/execution.rs @@ -8,6 +8,7 @@ //! execution. //! - `skill_front`: the skill front; compiles a skill run into an execution and seals it through the act engine. +pub(crate) mod disposition; pub(crate) mod fanout; pub(crate) mod graph; pub(crate) mod graph_index; diff --git a/crates/runx-runtime/src/execution/disposition.rs b/crates/runx-runtime/src/execution/disposition.rs new file mode 100644 index 000000000..1fd980f1f --- /dev/null +++ b/crates/runx-runtime/src/execution/disposition.rs @@ -0,0 +1,60 @@ +use runx_contracts::{ClosureDisposition, JsonValue}; +use thiserror::Error; + +#[derive(Debug, Error, Clone, PartialEq, Eq)] +pub(crate) enum ClosureDispositionParseError { + #[error("agent answer closure must be an object")] + ClosureNotObject, + #[error("agent answer closure.disposition is required")] + MissingDisposition, + #[error("agent answer closure.disposition must be a string")] + DispositionNotString, + #[error("agent answer closure.disposition {0:?} is not supported")] + UnsupportedDisposition(String), +} + +pub(crate) fn parse_agent_answer_disposition( + answer: &JsonValue, +) -> Result { + let closure = answer + .as_object() + .and_then(|object| object.get("closure")) + .ok_or(ClosureDispositionParseError::MissingDisposition)?; + let closure = closure + .as_object() + .ok_or(ClosureDispositionParseError::ClosureNotObject)?; + let disposition = closure + .get("disposition") + .ok_or(ClosureDispositionParseError::MissingDisposition)?; + let disposition = disposition + .as_str() + .ok_or(ClosureDispositionParseError::DispositionNotString)?; + parse_closure_disposition(disposition) +} + +pub(crate) fn agent_answer_disposition_or_closed( + answer: &JsonValue, +) -> Result { + match answer.as_object() { + Some(object) if object.contains_key("closure") => parse_agent_answer_disposition(answer), + _ => Ok(ClosureDisposition::Closed), + } +} + +pub(crate) fn parse_closure_disposition( + disposition: &str, +) -> Result { + match disposition { + "closed" => Ok(ClosureDisposition::Closed), + "deferred" => Ok(ClosureDisposition::Deferred), + "superseded" => Ok(ClosureDisposition::Superseded), + "declined" => Ok(ClosureDisposition::Declined), + "blocked" => Ok(ClosureDisposition::Blocked), + "failed" => Ok(ClosureDisposition::Failed), + "killed" => Ok(ClosureDisposition::Killed), + "timed_out" => Ok(ClosureDisposition::TimedOut), + other => Err(ClosureDispositionParseError::UnsupportedDisposition( + other.to_owned(), + )), + } +} diff --git a/crates/runx-runtime/src/execution/harness/runner.rs b/crates/runx-runtime/src/execution/harness/runner.rs index d0448d43e..115c8e1bc 100644 --- a/crates/runx-runtime/src/execution/harness/runner.rs +++ b/crates/runx-runtime/src/execution/harness/runner.rs @@ -582,7 +582,7 @@ fn replay_agent_skill_fixture( context: format!("serializing replay answer {request_id}"), source, })?; - let disposition = agent_answer_disposition(answer); + let disposition = agent_answer_disposition(answer)?; let succeeded = disposition == ClosureDisposition::Closed; Ok(( SkillOutput { diff --git a/crates/runx-runtime/src/execution/harness/runner/dispositions.rs b/crates/runx-runtime/src/execution/harness/runner/dispositions.rs index f866d00b7..f2d04fd4c 100644 --- a/crates/runx-runtime/src/execution/harness/runner/dispositions.rs +++ b/crates/runx-runtime/src/execution/harness/runner/dispositions.rs @@ -7,6 +7,7 @@ use super::super::super::super::adapter::{InvocationStatus, SkillOutput}; use super::super::fixtures::{HarnessExpectedStatus, HarnessFixture}; use super::HarnessReplayError; use crate::RuntimeError; +use crate::execution::disposition::agent_answer_disposition_or_closed; pub(super) fn agent_task_output( fixture: &HarnessFixture, @@ -80,23 +81,15 @@ pub(super) fn required_string_metadata( } } -pub(super) fn agent_answer_disposition(answer: &JsonValue) -> ClosureDisposition { - match answer - .as_object() - .and_then(|object| object.get("closure")) - .and_then(JsonValue::as_object) - .and_then(|closure| closure.get("disposition")) - .and_then(JsonValue::as_str) - { - Some("deferred") => ClosureDisposition::Deferred, - Some("superseded") => ClosureDisposition::Superseded, - Some("declined") => ClosureDisposition::Declined, - Some("blocked") => ClosureDisposition::Blocked, - Some("failed") => ClosureDisposition::Failed, - Some("killed") => ClosureDisposition::Killed, - Some("timed_out") => ClosureDisposition::TimedOut, - _ => ClosureDisposition::Closed, - } +pub(super) fn agent_answer_disposition( + answer: &JsonValue, +) -> Result { + agent_answer_disposition_or_closed(answer).map_err(|error| { + HarnessReplayError::InvalidReplayMetadata { + field: "caller.answers.*.closure.disposition".to_owned(), + message: error.to_string(), + } + }) } pub(super) fn disposition_from_expected_status( diff --git a/crates/runx-runtime/src/execution/orchestrator.rs b/crates/runx-runtime/src/execution/orchestrator.rs index 71ff92592..8192f95c2 100644 --- a/crates/runx-runtime/src/execution/orchestrator.rs +++ b/crates/runx-runtime/src/execution/orchestrator.rs @@ -90,6 +90,7 @@ pub struct HarnessRunRequest { pub struct InlineHarnessRequest { pub skill_path: PathBuf, pub receipt_dir: Option, + pub env: Option>, } #[derive(Clone, Debug, Default, PartialEq, Eq)] @@ -205,6 +206,7 @@ impl LocalOrchestrator { Ok(super::skill_front::run_inline_harness_with_effects( &request.skill_path, request.receipt_dir.as_deref(), + request.env.as_ref(), &self.effects, )?) } diff --git a/crates/runx-runtime/src/execution/runner/steps.rs b/crates/runx-runtime/src/execution/runner/steps.rs index e36febdf4..13e160d5b 100644 --- a/crates/runx-runtime/src/execution/runner/steps.rs +++ b/crates/runx-runtime/src/execution/runner/steps.rs @@ -37,6 +37,7 @@ use crate::agent_invocation::{ }; use crate::approval::ApprovalResolution; use crate::effects::EffectReplay; +use crate::execution::disposition::agent_answer_disposition_or_closed; use crate::execution::output_projection::{StepOutputProjection, project_step_output}; use crate::host::Host; use crate::receipts::{StepSeal, StepSealClosure, seal_step}; @@ -754,7 +755,12 @@ fn replay_skill_output( reason: "effect replay output status must be a string".to_owned(), }); } - None => InvocationStatus::Success, + None => { + return Err(RuntimeError::InvalidRunStep { + step_id: step.id.clone(), + reason: "effect replay output status is required".to_owned(), + }); + } }; let stdout = match outputs.get("stdout") { Some(JsonValue::String(value)) => value.clone(), @@ -950,17 +956,7 @@ fn cli_tool_source(step: &GraphStep) -> Result { step_id: step.id.clone(), reason: "run.command is required for a cli-tool step".to_owned(), })?; - let args = run - .get("args") - .and_then(JsonValue::as_array) - .map(|values| { - values - .iter() - .filter_map(JsonValue::as_str) - .map(str::to_owned) - .collect::>() - }) - .unwrap_or_default(); + let args = cli_tool_args(step, run)?; Ok(SkillSource { act: None, source_type: SourceKind::CliTool, @@ -986,6 +982,29 @@ fn cli_tool_source(step: &GraphStep) -> Result { }) } +fn cli_tool_args(step: &GraphStep, run: &JsonObject) -> Result, RuntimeError> { + let Some(value) = run.get("args") else { + return Ok(Vec::new()); + }; + let JsonValue::Array(values) = value else { + return Err(RuntimeError::InvalidRunStep { + step_id: step.id.clone(), + reason: "run.args must be an array".to_owned(), + }); + }; + values + .iter() + .enumerate() + .map(|(index, value)| match value { + JsonValue::String(arg) => Ok(arg.clone()), + _ => Err(RuntimeError::InvalidRunStep { + step_id: step.id.clone(), + reason: format!("run.args[{index}] must be a string"), + }), + }) + .collect() +} + // The shared close for an agent act: a resolved host response becomes the // step's output, projection, and sealed receipt. Both the inline `agent-task` // step and a referenced agent skill end here, so the agent-act seal lives in @@ -998,8 +1017,8 @@ fn seal_agent_act_step( skill_name: String, response: ResolutionResponse, ) -> Result { - let disposition = agent_answer_disposition_value(&response.payload); - let output = agent_task_output(response)?; + let disposition = agent_answer_disposition_value(step, &response.payload)?; + let output = agent_task_output(response, &disposition)?; let projection = build_step_output_projection(step, &output, ClaimContextExposure::DeclaredOnly)?; let disposition_label = closure_disposition_label(&disposition); @@ -1310,9 +1329,11 @@ fn catalog_source(tool_ref: &str) -> SkillSource { } } -fn agent_task_output(response: ResolutionResponse) -> Result { - let disposition = agent_answer_disposition_value(&response.payload); - let succeeded = disposition == ClosureDisposition::Closed; +fn agent_task_output( + response: ResolutionResponse, + disposition: &ClosureDisposition, +) -> Result { + let succeeded = *disposition == ClosureDisposition::Closed; let stdout = serde_json::to_string(&response.payload) .map_err(|source| RuntimeError::json("serializing agent-task response", source))?; Ok(SkillOutput { @@ -1327,7 +1348,7 @@ fn agent_task_output(response: ResolutionResponse) -> Result Option { } } -fn agent_answer_disposition_value(answer: &JsonValue) -> ClosureDisposition { - match answer - .as_object() - .and_then(|object| object.get("closure")) - .and_then(JsonValue::as_object) - .and_then(|closure| closure.get("disposition")) - .and_then(JsonValue::as_str) - { - Some("deferred") => ClosureDisposition::Deferred, - Some("superseded") => ClosureDisposition::Superseded, - Some("declined") => ClosureDisposition::Declined, - Some("blocked") => ClosureDisposition::Blocked, - Some("failed") => ClosureDisposition::Failed, - Some("killed") => ClosureDisposition::Killed, - Some("timed_out") => ClosureDisposition::TimedOut, - _ => ClosureDisposition::Closed, - } +fn agent_answer_disposition_value( + step: &GraphStep, + answer: &JsonValue, +) -> Result { + agent_answer_disposition_or_closed(answer).map_err(|error| RuntimeError::InvalidRunStep { + step_id: step.id.clone(), + reason: format!("{error}"), + }) } fn closure_disposition_label(disposition: &ClosureDisposition) -> &'static str { diff --git a/crates/runx-runtime/src/execution/runner/steps/output.rs b/crates/runx-runtime/src/execution/runner/steps/output.rs index 85d40e23c..2d22187cc 100644 --- a/crates/runx-runtime/src/execution/runner/steps/output.rs +++ b/crates/runx-runtime/src/execution/runner/steps/output.rs @@ -46,14 +46,13 @@ fn expose_declared_run_outputs( let Some(JsonValue::Object(declared_outputs)) = run.get("outputs") else { return Ok(()); }; - if claim.is_empty() { - return Ok(()); - } - for name in declared_outputs.keys() { reject_reserved_step_output_name(step, name, "declared run output")?; let Some(value) = declared_claim_value(claim, name) else { - continue; + return Err(RuntimeError::InvalidRunStep { + step_id: step.id.clone(), + reason: format!("declared run output {name:?} was not returned by the step"), + }); }; outputs.insert(name.clone(), value); } diff --git a/crates/runx-runtime/src/execution/skill_front.rs b/crates/runx-runtime/src/execution/skill_front.rs index 10068cf15..31eb1a721 100644 --- a/crates/runx-runtime/src/execution/skill_front.rs +++ b/crates/runx-runtime/src/execution/skill_front.rs @@ -18,6 +18,7 @@ use crate::RuntimeError; use crate::adapter::{InvocationStatus, SkillInvocation, SkillOutput}; use crate::agent_invocation::{AgentActInvocationSourceType, agent_act_resolution_request}; use crate::effects::RuntimeEffectRegistry; +use crate::execution::disposition::agent_answer_disposition_or_closed; use crate::execution::orchestrator::SkillRunRequest; use crate::execution::output_projection::project_step_output; use crate::receipts::signing::strip_receipt_signing_env; @@ -355,29 +356,50 @@ fn build_domain_act_frame( .and_then(map_decision_choice) .unwrap_or(DecisionChoice::Close); - // The effect ref: a venue id read from the real governed tool result (never - // the model's restatement), wrapped into a domain uri. e.g. the `/v1` - // response's `id` becomes `frantic:judgment:` for the venue to reconcile. - let artifact_refs = governed_effect + let reference_type = match act.effect_type.as_deref().unwrap_or("artifact") { + "act" => ReferenceType::Act, + "tracking_item" => ReferenceType::TrackingItem, + "receipt" => ReferenceType::Receipt, + "provider" | "provider_event" => ReferenceType::ProviderEvent, + "provider_thread" => ReferenceType::ProviderThread, + "provider_comment" => ReferenceType::ProviderComment, + "github_issue" => ReferenceType::GithubIssue, + "external_url" => ReferenceType::ExternalUrl, + _ => ReferenceType::Artifact, + }; + let prefix = resolve( + act.effect_prefix_from.as_deref(), + act.effect_prefix.as_deref(), + ) + .unwrap_or_default(); + let effect_ref = |id: &str| { + let id = id.trim(); + (!id.is_empty()) + .then(|| Reference::with_uri(reference_type.clone(), format!("{prefix}{id}"))) + }; + let mut artifact_refs = Vec::new(); + if let Some(reference) = act + .effect_from_input + .as_deref() + .and_then(|key| inputs.get(key)) + .and_then(JsonValue::as_str) + .and_then(effect_ref) + { + artifact_refs.push(reference); + } + if let Some(reference) = governed_effect .and_then(|effect| { - let field = resolve(None, act.effect_from.as_deref())?; - let id = effect + let field = resolve(act.effect_field_from.as_deref(), act.effect_from.as_deref())?; + effect .as_object() .and_then(|object| object.get(field.as_str())) .and_then(JsonValue::as_str) - .map(str::trim) - .filter(|value| !value.is_empty())?; - let reference_type = match act.effect_type.as_deref().unwrap_or("artifact") { - "act" => ReferenceType::Act, - "tracking_item" => ReferenceType::TrackingItem, - "receipt" => ReferenceType::Receipt, - _ => ReferenceType::Artifact, - }; - let prefix = resolve(None, act.effect_prefix.as_deref()).unwrap_or_default(); - Some(Reference::with_uri(reference_type, format!("{prefix}{id}"))) + .and_then(effect_ref) }) - .into_iter() - .collect::>(); + .filter(|reference| !artifact_refs.contains(reference)) + { + artifact_refs.push(reference); + } Some(DomainActFrame { form, @@ -448,23 +470,8 @@ fn seal_skill_output( )?) } -fn answer_disposition(answer: &JsonValue) -> ClosureDisposition { - match answer - .as_object() - .and_then(|object| object.get("closure")) - .and_then(JsonValue::as_object) - .and_then(|closure| closure.get("disposition")) - .and_then(JsonValue::as_str) - { - Some("deferred") => ClosureDisposition::Deferred, - Some("superseded") => ClosureDisposition::Superseded, - Some("declined") => ClosureDisposition::Declined, - Some("blocked") => ClosureDisposition::Blocked, - Some("failed") => ClosureDisposition::Failed, - Some("killed") => ClosureDisposition::Killed, - Some("timed_out") => ClosureDisposition::TimedOut, - _ => ClosureDisposition::Closed, - } +fn answer_disposition(answer: &JsonValue) -> Result { + agent_answer_disposition_or_closed(answer).map_err(|error| invalid(format!("{error}"))) } fn sealed_output( diff --git a/crates/runx-runtime/src/execution/skill_front/agent.rs b/crates/runx-runtime/src/execution/skill_front/agent.rs index 25e696cd0..b6a27cb81 100644 --- a/crates/runx-runtime/src/execution/skill_front/agent.rs +++ b/crates/runx-runtime/src/execution/skill_front/agent.rs @@ -57,7 +57,7 @@ pub(super) fn execute_agent_skill_run( }; let stdout = serde_json::to_string(&answer) .map_err(|error| SkillRunError::Invalid(format!("failed to serialize answer: {error}")))?; - let disposition = answer_disposition(&answer); + let disposition = answer_disposition(&answer)?; let receipt = match domain_act_frame(&invocation, &answer, governed_effect.as_ref()) { Some(frame) => { let label = closure_disposition_label(&disposition); @@ -148,7 +148,7 @@ fn try_inline_agent_resolution( id: agent_act.id.clone(), invocation: Box::new(agent_act), }; - let transport = ReqwestHttpTransport::new().map_err(|error| { + let transport = ReqwestHttpTransport::for_managed_agent().map_err(|error| { SkillRunError::Invalid(format!("managed agent transport error: {error}")) })?; let resolver = AnthropicAgentResolver::new( diff --git a/crates/runx-runtime/src/execution/skill_front/graph.rs b/crates/runx-runtime/src/execution/skill_front/graph.rs index b881a38f3..10eae4401 100644 --- a/crates/runx-runtime/src/execution/skill_front/graph.rs +++ b/crates/runx-runtime/src/execution/skill_front/graph.rs @@ -6,6 +6,7 @@ use super::{ contract_json_value, identifier_segment, invalid, needs_agent_output, sealed_output, }; +use std::collections::BTreeMap; use std::path::PathBuf; use runx_contracts::{ @@ -73,6 +74,7 @@ pub(super) fn execute_graph_skill_run( credential_delivery_from_invocation(workspace.env(), request.local_credential.as_ref())?; let inline_resolver = InlineResolver { skill_directory: skill_dir.clone(), + env: env.clone(), credential_delivery: credential_delivery.clone(), }; let runtime = Runtime::new( @@ -438,6 +440,8 @@ struct InlineResolver { #[cfg_attr(not(feature = "agent"), allow(dead_code))] skill_directory: PathBuf, #[cfg_attr(not(feature = "agent"), allow(dead_code))] + env: BTreeMap, + #[cfg_attr(not(feature = "agent"), allow(dead_code))] credential_delivery: CredentialDelivery, } @@ -452,23 +456,22 @@ impl InlineResolver { skill_name: "managed-agent".to_owned(), message, }; - // The same process-env snapshot the rest of the runtime reads, so the - // inline graph agent path resolves the provider exactly like the - // top-level agent path rather than reaching for raw `std::env`. - let env = crate::services::process_env_snapshot(); - let config = match crate::config::load_managed_agent_config(&env, &self.skill_directory) - .map_err(|error| fail(format!("managed agent config error: {error}")))? - { - Some(config) if config.provider.as_str().eq_ignore_ascii_case("anthropic") => config, - _ => return Ok(None), - }; - let transport = ReqwestHttpTransport::new() + let config = + match crate::config::load_managed_agent_config(&self.env, &self.skill_directory) + .map_err(|error| fail(format!("managed agent config error: {error}")))? + { + Some(config) if config.provider.as_str().eq_ignore_ascii_case("anthropic") => { + config + } + _ => return Ok(None), + }; + let transport = ReqwestHttpTransport::for_managed_agent() .map_err(|error| fail(format!("managed agent transport error: {error}")))?; let resolver = AnthropicAgentResolver::new( transport, config.api_key, config.model, - env, + self.env.clone(), self.skill_directory.clone(), self.credential_delivery.clone(), ); diff --git a/crates/runx-runtime/src/execution/skill_front/inline_harness.rs b/crates/runx-runtime/src/execution/skill_front/inline_harness.rs index 6684696e0..e3ba24f11 100644 --- a/crates/runx-runtime/src/execution/skill_front/inline_harness.rs +++ b/crates/runx-runtime/src/execution/skill_front/inline_harness.rs @@ -23,6 +23,7 @@ use super::runner_manifest::{load_runner_manifest, resolve_skill_dir, selected_r pub(crate) fn run_inline_harness_with_effects( skill_path: &Path, receipt_dir: Option<&Path>, + env: Option<&BTreeMap>, effects: &RuntimeEffectRegistry, ) -> Result { let skill_dir = resolve_skill_dir(skill_path)?; @@ -45,7 +46,7 @@ pub(crate) fn run_inline_harness_with_effects( for case in &harness.cases { case_names.push(case.name.clone()); let outcome = - run_inline_harness_case(&skill_dir, receipt_dir, &manifest, case, &cwd, effects); + run_inline_harness_case(&skill_dir, receipt_dir, env, &manifest, case, &cwd, effects); if outcome.is_graph { graph_case_count += 1; } @@ -82,6 +83,7 @@ struct InlineHarnessCaseOutcome { fn run_inline_harness_case( skill_dir: &Path, receipt_dir: Option<&Path>, + env: Option<&BTreeMap>, manifest: &SkillRunnerManifest, case: &RunnerHarnessCase, cwd: &Path, @@ -91,7 +93,7 @@ fn run_inline_harness_case( Ok(runner) => runner.source.source_type == runx_parser::SourceKind::Graph, Err(error) => return inline_harness_case_error(&case.name, error), }; - let request = inline_harness_case_request(skill_dir, receipt_dir, case, cwd); + let request = inline_harness_case_request(skill_dir, receipt_dir, env, case, cwd); let overrides = SkillRunOverrides { runner: case.runner.clone(), seeded_answers: seeded_answers_from_caller(&case.caller), @@ -113,10 +115,12 @@ fn run_inline_harness_case( fn inline_harness_case_request( skill_dir: &Path, receipt_dir: Option<&Path>, + env: Option<&BTreeMap>, case: &RunnerHarnessCase, cwd: &Path, ) -> SkillRunRequest { - let mut env: BTreeMap = std::env::vars().collect(); + let mut env: BTreeMap = + env.cloned().unwrap_or_else(|| std::env::vars().collect()); env.extend(case.env.clone()); SkillRunRequest { skill_path: skill_dir.to_path_buf(), diff --git a/crates/runx-runtime/src/registry/local.rs b/crates/runx-runtime/src/registry/local.rs index 74c6d1f7e..55725fcd3 100644 --- a/crates/runx-runtime/src/registry/local.rs +++ b/crates/runx-runtime/src/registry/local.rs @@ -90,6 +90,8 @@ pub enum LocalRegistryError { }, #[error("invalid registry version payload at {field}: {message}")] InvalidVersionPayload { field: String, message: String }, + #[error("invalid registry skill manifest at {field}: {message}")] + InvalidSkillManifest { field: String, message: String }, #[error("invalid registry skill id '{0}'. Expected '/'.")] InvalidSkillId(String), #[error("registry slugs cannot be empty")] @@ -214,6 +216,8 @@ impl FileRegistryStore { owner: latest.owner.clone(), name: latest.name.clone(), description: latest.description.clone(), + category: latest.category.clone(), + source_category: latest.source_category.clone(), latest_version: latest.version.clone(), latest_digest: latest.digest.clone(), versions, diff --git a/crates/runx-runtime/src/registry/local/build.rs b/crates/runx-runtime/src/registry/local/build.rs index c99de9f1b..5de87f4b0 100644 --- a/crates/runx-runtime/src/registry/local/build.rs +++ b/crates/runx-runtime/src/registry/local/build.rs @@ -48,6 +48,8 @@ pub fn build_registry_skill_version( owner: defaults.owner, name: skill.name.clone(), description: skill.description.clone(), + category: skill.runx_category.clone(), + source_category: skill.category.clone(), version: defaults.version, digest, signed_manifest: None, @@ -67,7 +69,7 @@ pub fn build_registry_skill_version( catalog_visibility: Some(catalog.visibility.as_str().to_owned()), source_metadata: defaults.source_metadata, attestations: defaults.attestations, - required_scopes: registry_required_scopes(&skill, manifest), + required_scopes: registry_required_scopes(&skill, manifest)?, runtime: registry_runtime(&skill, manifest), auth: skill.auth.clone(), risk: registry_risk(&skill), @@ -146,8 +148,13 @@ pub(super) fn registry_catalog( pub(super) fn registry_required_scopes( skill: &ValidatedSkill, manifest: Option<&SkillRunnerManifest>, -) -> Vec { - required_scopes_from_skill_and_runner(skill, manifest) +) -> Result, LocalRegistryError> { + required_scopes_from_skill_and_runner(skill, manifest).map_err(|error| { + LocalRegistryError::InvalidSkillManifest { + field: error.field, + message: error.message, + } + }) } pub(super) fn registry_runtime( @@ -175,11 +182,14 @@ pub(super) fn registry_tags( unique( extract_tags(skill) .into_iter() + .chain(skill.runx_category.clone()) .chain(extract_runner_tags(manifest)) .collect(), ) } +// rust-style-allow: long-function - normalization validates the package digest, +// manifest, and registry row in one pass over the submitted version payload. pub fn normalize_registry_skill_version( payload: RegistrySkillVersionPayload, ) -> Result { @@ -202,15 +212,22 @@ pub fn normalize_registry_skill_version( message: "declared without package_files".to_owned(), }); } + let markdown = required_string(payload.markdown, "registry_version.markdown")?; + let derived_categories = derive_categories_from_markdown(&markdown); + let category = payload.category.or(derived_categories.runx_category); + let source_category = payload.source_category.or(derived_categories.category); + Ok(RegistrySkillVersion { skill_id: required_string(payload.skill_id, "registry_version.skill_id")?, owner: governance.owner, name: required_string(payload.name, "registry_version.name")?, description: payload.description, + category, + source_category, version: required_string(payload.version, "registry_version.version")?, digest: required_string(payload.digest, "registry_version.digest")?, signed_manifest: payload.signed_manifest, - markdown: required_string(payload.markdown, "registry_version.markdown")?, + markdown, profile_document: payload.profile_document, profile_digest: payload.profile_digest, package_files, @@ -237,6 +254,27 @@ pub fn normalize_registry_skill_version( }) } +struct DerivedCategories { + category: Option, + runx_category: Option, +} + +fn derive_categories_from_markdown(markdown: &str) -> DerivedCategories { + let Some(skill) = parse_skill_markdown(markdown) + .ok() + .and_then(|raw| validate_skill(raw).ok()) + else { + return DerivedCategories { + category: None, + runx_category: None, + }; + }; + DerivedCategories { + category: skill.category, + runx_category: skill.runx_category, + } +} + struct NormalizedRegistryVersionGovernance { owner: String, created_at: String, @@ -311,6 +349,7 @@ pub(super) fn normalize_registry_catalog( audience: match audience { Some("builder") => runx_parser::CatalogAudience::Builder, Some("operator") => runx_parser::CatalogAudience::Operator, + Some("system") => runx_parser::CatalogAudience::System, _ => runx_parser::CatalogAudience::Public, }, visibility: match visibility { @@ -331,6 +370,8 @@ pub struct RegistrySkillVersionPayload { owner: Option, name: Option, description: Option, + category: Option, + source_category: Option, version: Option, digest: Option, signed_manifest: Option, diff --git a/crates/runx-runtime/src/registry/local/trust.rs b/crates/runx-runtime/src/registry/local/trust.rs index aa078e280..063588f1c 100644 --- a/crates/runx-runtime/src/registry/local/trust.rs +++ b/crates/runx-runtime/src/registry/local/trust.rs @@ -303,6 +303,8 @@ pub(super) fn search_result_for_version( skill_id: version.skill_id.clone(), name: version.name.clone(), summary: version.description.clone(), + category: version.category.clone(), + source_category: version.source_category.clone(), owner: version.owner.clone(), version: Some(version.version.clone()), digest: Some(version.digest.clone()), @@ -339,6 +341,8 @@ pub(super) fn detail_for_version( owner: version.owner.clone(), name: version.name.clone(), description: version.description.clone(), + category: version.category.clone(), + source_category: version.source_category.clone(), version: version.version.clone(), digest: version.digest.clone(), signed_manifest: version.signed_manifest.clone(), @@ -389,6 +393,12 @@ pub(super) fn searchable_text(version: &RegistrySkillVersion) -> String { if let Some(description) = &version.description { parts.push(description.clone()); } + if let Some(category) = &version.category { + parts.push(category.clone()); + } + if let Some(source_category) = &version.source_category { + parts.push(source_category.clone()); + } parts.extend(version.runner_names.clone()); parts.extend(version.tags.clone()); normalize(&parts.join(" ")) diff --git a/crates/runx-runtime/src/registry/payload.rs b/crates/runx-runtime/src/registry/payload.rs index fd71357c9..f720ca380 100644 --- a/crates/runx-runtime/src/registry/payload.rs +++ b/crates/runx-runtime/src/registry/payload.rs @@ -26,6 +26,8 @@ pub(crate) fn parse_search( skill_id: string_field(skill, "skill_id", route, &path)?, name: string_field(skill, "name", route, &path)?, summary: optional_string_field(skill, "description", route, &path)?, + category: optional_string_field(skill, "category", route, &path)?, + source_category: optional_string_field(skill, "source_category", route, &path)?, owner: string_field(skill, "owner", route, &path)?, version: optional_string_field(skill, "version", route, &path)?, digest: optional_string_field(skill, "digest", route, &path)?, @@ -64,6 +66,8 @@ pub(crate) fn parse_read( owner: string_field(skill, "owner", route, "$.skill")?, name: string_field(skill, "name", route, "$.skill")?, description: optional_string_field(skill, "description", route, "$.skill")?, + category: optional_string_field(skill, "category", route, "$.skill")?, + source_category: optional_string_field(skill, "source_category", route, "$.skill")?, version: string_field(skill, "version", route, "$.skill")?, digest: string_field(skill, "digest", route, "$.skill")?, signed_manifest: signed_manifest_field(skill, "signed_manifest", route, "$.skill")?, diff --git a/crates/runx-runtime/src/registry/scopes.rs b/crates/runx-runtime/src/registry/scopes.rs index bbe623c31..0f5ebb4a1 100644 --- a/crates/runx-runtime/src/registry/scopes.rs +++ b/crates/runx-runtime/src/registry/scopes.rs @@ -1,65 +1,116 @@ use runx_contracts::{JsonObject, JsonValue}; use runx_parser::{SkillRunnerManifest, ValidatedSkill}; -pub(crate) fn required_scopes_from_skill(skill: &ValidatedSkill) -> Vec { - unique_strings( - string_array_field(skill.auth.as_ref(), "scopes") +#[derive(Clone, Debug, PartialEq, Eq)] +pub(crate) struct ScopeParseError { + pub(crate) field: String, + pub(crate) message: String, +} + +impl std::fmt::Display for ScopeParseError { + fn fmt(&self, formatter: &mut std::fmt::Formatter<'_>) -> std::fmt::Result { + write!(formatter, "{} {}", self.field, self.message) + } +} + +impl std::error::Error for ScopeParseError {} + +pub(crate) fn required_scopes_from_skill( + skill: &ValidatedSkill, +) -> Result, ScopeParseError> { + Ok(unique_strings( + string_array_field(skill.auth.as_ref(), "auth.scopes")? .into_iter() .chain(string_array_field_from_object( skill.runx.as_ref(), - "scopes", - )), - ) + "runx.scopes", + )?), + )) } pub(super) fn required_scopes_from_skill_and_runner( skill: &ValidatedSkill, manifest: Option<&SkillRunnerManifest>, -) -> Vec { - unique_strings( - required_scopes_from_skill(skill) +) -> Result, ScopeParseError> { + Ok(unique_strings( + required_scopes_from_skill(skill)? .into_iter() - .chain(required_scopes_from_runner_manifest(manifest)), - ) + .chain(required_scopes_from_runner_manifest(manifest)?), + )) } -fn required_scopes_from_runner_manifest(manifest: Option<&SkillRunnerManifest>) -> Vec { - unique_strings( - manifest - .into_iter() - .flat_map(|manifest| manifest.runners.values()) - .flat_map(|runner| { - string_array_field(runner.auth.as_ref(), "scopes") - .into_iter() - .chain(string_array_field_from_object( - runner.runx.as_ref(), - "scopes", - )) - }), - ) +fn required_scopes_from_runner_manifest( + manifest: Option<&SkillRunnerManifest>, +) -> Result, ScopeParseError> { + let mut scopes = Vec::new(); + let Some(manifest) = manifest else { + return Ok(scopes); + }; + for (runner_name, runner) in &manifest.runners { + scopes.extend(string_array_field( + runner.auth.as_ref(), + &format!("runners.{runner_name}.auth.scopes"), + )?); + scopes.extend(string_array_field_from_object( + runner.runx.as_ref(), + &format!("runners.{runner_name}.runx.scopes"), + )?); + } + Ok(unique_strings(scopes)) } -fn string_array_field(value: Option<&JsonValue>, field: &str) -> Vec { - let Some(JsonValue::Object(record)) = value else { - return Vec::new(); +fn string_array_field( + value: Option<&JsonValue>, + field: &str, +) -> Result, ScopeParseError> { + let Some(value) = value else { + return Ok(Vec::new()); + }; + let JsonValue::Object(record) = value else { + return Err(ScopeParseError { + field: field_parent(field).to_owned(), + message: "must be an object when declaring scopes".to_owned(), + }); }; string_array_field_from_object(Some(record), field) } -fn string_array_field_from_object(value: Option<&JsonObject>, field: &str) -> Vec { +fn string_array_field_from_object( + value: Option<&JsonObject>, + field: &str, +) -> Result, ScopeParseError> { let Some(record) = value else { - return Vec::new(); + return Ok(Vec::new()); }; - let Some(JsonValue::Array(values)) = record.get(field) else { - return Vec::new(); + let scope_key = field.rsplit('.').next().unwrap_or(field); + let Some(value) = record.get(scope_key) else { + return Ok(Vec::new()); }; - values - .iter() - .filter_map(JsonValue::as_str) - .map(str::trim) - .filter(|scope| !scope.is_empty()) - .map(str::to_owned) - .collect() + let JsonValue::Array(values) = value else { + return Err(ScopeParseError { + field: field.to_owned(), + message: "must be an array of non-empty strings".to_owned(), + }); + }; + let mut scopes = Vec::new(); + for (index, value) in values.iter().enumerate() { + let Some(scope) = value + .as_str() + .map(str::trim) + .filter(|scope| !scope.is_empty()) + else { + return Err(ScopeParseError { + field: format!("{field}[{index}]"), + message: "must be a non-empty string".to_owned(), + }); + }; + scopes.push(scope.to_owned()); + } + Ok(scopes) +} + +fn field_parent(field: &str) -> &str { + field.rsplit_once('.').map_or(field, |(parent, _)| parent) } fn unique_strings(values: impl IntoIterator) -> Vec { @@ -71,3 +122,47 @@ fn unique_strings(values: impl IntoIterator) -> Vec { } unique_values } + +#[cfg(test)] +mod tests { + use super::*; + + #[test] + fn scope_arrays_reject_non_string_entries() -> Result<(), Box> { + let mut record = JsonObject::new(); + record.insert( + "scopes".to_owned(), + JsonValue::Array(vec![ + JsonValue::String("repo.read".to_owned()), + JsonValue::Bool(true), + ]), + ); + + let error = match string_array_field_from_object(Some(&record), "auth.scopes") { + Ok(_) => return Err("malformed scope entry should fail closed".into()), + Err(error) => error, + }; + assert_eq!(error.field, "auth.scopes[1]"); + Ok(()) + } + + #[test] + fn scope_arrays_trim_deduplicate_and_reject_empty_entries() + -> Result<(), Box> { + let mut record = JsonObject::new(); + record.insert( + "scopes".to_owned(), + JsonValue::Array(vec![ + JsonValue::String(" repo.read ".to_owned()), + JsonValue::String("".to_owned()), + ]), + ); + + let error = match string_array_field_from_object(Some(&record), "auth.scopes") { + Ok(_) => return Err("empty scope entry should fail closed".into()), + Err(error) => error, + }; + assert_eq!(error.field, "auth.scopes[1]"); + Ok(()) + } +} diff --git a/crates/runx-runtime/src/registry/types.rs b/crates/runx-runtime/src/registry/types.rs index 8c6c4bed4..46a3a8736 100644 --- a/crates/runx-runtime/src/registry/types.rs +++ b/crates/runx-runtime/src/registry/types.rs @@ -124,6 +124,10 @@ pub struct RegistrySearchResult { pub name: String, #[serde(skip_serializing_if = "Option::is_none")] pub summary: Option, + #[serde(skip_serializing_if = "Option::is_none")] + pub category: Option, + #[serde(skip_serializing_if = "Option::is_none")] + pub source_category: Option, pub owner: String, #[serde(skip_serializing_if = "Option::is_none")] pub version: Option, @@ -156,6 +160,10 @@ pub struct RegistrySkillVersion { pub name: String, #[serde(skip_serializing_if = "Option::is_none")] pub description: Option, + #[serde(skip_serializing_if = "Option::is_none")] + pub category: Option, + #[serde(skip_serializing_if = "Option::is_none")] + pub source_category: Option, pub version: String, pub digest: String, #[serde(skip_serializing_if = "Option::is_none")] @@ -206,6 +214,10 @@ pub struct RegistrySkill { pub name: String, #[serde(skip_serializing_if = "Option::is_none")] pub description: Option, + #[serde(skip_serializing_if = "Option::is_none")] + pub category: Option, + #[serde(skip_serializing_if = "Option::is_none")] + pub source_category: Option, pub latest_version: String, pub latest_digest: String, pub versions: Vec, @@ -311,6 +323,10 @@ pub struct RegistrySkillDetail { pub name: String, #[serde(skip_serializing_if = "Option::is_none")] pub description: Option, + #[serde(skip_serializing_if = "Option::is_none")] + pub category: Option, + #[serde(skip_serializing_if = "Option::is_none")] + pub source_category: Option, pub version: String, pub digest: String, #[serde(skip_serializing_if = "Option::is_none")] diff --git a/crates/runx-runtime/src/runtime_http.rs b/crates/runx-runtime/src/runtime_http.rs index 8d255b31e..84d21a2a6 100644 --- a/crates/runx-runtime/src/runtime_http.rs +++ b/crates/runx-runtime/src/runtime_http.rs @@ -120,14 +120,72 @@ pub struct ReqwestHttpTransport { #[cfg(feature = "async-http")] const MAX_HTTP_RESPONSE_BYTES: usize = 1024 * 1024; +#[cfg(feature = "async-http")] +const DEFAULT_HTTP_REQUEST_TIMEOUT: Duration = Duration::from_secs(30); +#[cfg(feature = "async-http")] +const DEFAULT_HTTP_CONNECT_TIMEOUT: Duration = Duration::from_secs(10); +#[cfg(feature = "async-http")] +const MANAGED_AGENT_REQUEST_TIMEOUT: Duration = Duration::from_secs(180); + +/// The default browser User-Agent the governed fetch transport presents (current +/// stable Chrome). Overridable per run with `RUNX_HTTP_USER_AGENT`, opt-out with +/// `RUNX_HTTP_BROWSER=0`. This is header/UA-level emulation only: it clears basic bot +/// scoring (a missing UA, no browser headers), NOT TLS (JA3/JA4) or HTTP/2 +/// fingerprinting, which would need a Chrome-impersonating TLS stack and is +/// deliberately out of scope. Sites on a JS/managed challenge, or that fingerprint the +/// rustls handshake, are expected to still block us; we surface that as a non-2xx +/// rather than escalate. +#[allow(dead_code)] // consumed by the feature-gated http adapter and the transport tests +pub const DEFAULT_BROWSER_USER_AGENT: &str = "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/143.0.0.0 Safari/537.36"; + +/// The Chrome navigation header set, applied as client default headers so a per-request +/// (manifest/caller) header of the same name still overrides it. The User-Agent is set +/// via the builder's `.user_agent()` and Accept-Encoding is owned by the gzip/brotli/... +/// decoders, so neither is here. reqwest's HeaderMap is hash-ordered and will not +/// reproduce Chrome's header order on the wire: values match Chrome, order does not, +/// which is the honest ceiling for header-level emulation. +#[cfg(feature = "async-http")] +fn chrome_default_headers() -> reqwest::header::HeaderMap { + use reqwest::header::{HeaderMap, HeaderValue}; + let mut headers = HeaderMap::new(); + headers.insert( + "sec-ch-ua", + HeaderValue::from_static( + "\"Google Chrome\";v=\"143\", \"Chromium\";v=\"143\", \"Not/A)Brand\";v=\"24\"", + ), + ); + headers.insert("sec-ch-ua-mobile", HeaderValue::from_static("?0")); + headers.insert( + "sec-ch-ua-platform", + HeaderValue::from_static("\"Windows\""), + ); + headers.insert("upgrade-insecure-requests", HeaderValue::from_static("1")); + headers.insert( + "accept", + HeaderValue::from_static( + "text/html,application/xhtml+xml,application/xml;q=0.9,image/avif,image/webp,image/apng,*/*;q=0.8,application/signed-exchange;v=b3;q=0.7", + ), + ); + headers.insert("sec-fetch-site", HeaderValue::from_static("none")); + headers.insert("sec-fetch-mode", HeaderValue::from_static("navigate")); + headers.insert("sec-fetch-user", HeaderValue::from_static("?1")); + headers.insert("sec-fetch-dest", HeaderValue::from_static("document")); + headers.insert( + "accept-language", + HeaderValue::from_static("en-US,en;q=0.9"), + ); + headers.insert("priority", HeaderValue::from_static("u=0, i")); + headers +} #[cfg(feature = "async-http")] impl ReqwestHttpTransport { pub fn new() -> Result { Self::with_timeouts_and_private_networks( - Duration::from_secs(30), - Duration::from_secs(10), + DEFAULT_HTTP_REQUEST_TIMEOUT, + DEFAULT_HTTP_CONNECT_TIMEOUT, false, + None, ) } @@ -135,22 +193,40 @@ impl ReqwestHttpTransport { request_timeout: Duration, connect_timeout: Duration, allow_private_networks: bool, + browser_user_agent: Option, ) -> Result { // reqwest is built with `rustls-no-provider`, so the process needs a // default crypto provider before a TLS client can be constructed. // Install ring once; an Err means another transport already set it. let _ = rustls::crypto::ring::default_provider().install_default(); + // Decode like a browser (the decoders also advertise the matching + // Accept-Encoding) and let ALPN negotiate HTTP/2; a no-compression, + // http1-only client is a bot tell. The response cap measures DECODED + // bytes (read_limited_response_body), so a decompression bomb stays bounded. let mut builder = reqwest::Client::builder() .redirect(reqwest::redirect::Policy::none()) .timeout(request_timeout) - .connect_timeout(connect_timeout); + .connect_timeout(connect_timeout) + .gzip(true) + .brotli(true) + .deflate(true) + .zstd(true); + // The browser profile is a default-header layer: a per-request + // (manifest/caller) header of the same name still overrides it. The UA goes + // through the dedicated builder method so a caller UA header overrides it + // without duplicating. None = the plain client (internal/API callers). + if let Some(user_agent) = browser_user_agent { + builder = builder + .user_agent(user_agent) + .default_headers(chrome_default_headers()); + } if !allow_private_networks { builder = builder.dns_resolver(GuardedDnsResolver::new(TokioDnsResolver)); } let client = builder .build() .map_err(|error| RuntimeHttpError::Transport { - message: error.to_string(), + message: transport_error_message(&error), })?; Ok(Self { client, @@ -164,9 +240,40 @@ impl ReqwestHttpTransport { /// `allowPrivateNetwork`) before choosing it, never as a default. pub fn with_private_network_access() -> Result { Self::with_timeouts_and_private_networks( - Duration::from_secs(30), - Duration::from_secs(10), + DEFAULT_HTTP_REQUEST_TIMEOUT, + DEFAULT_HTTP_CONNECT_TIMEOUT, true, + None, + ) + } + + /// Build the model-provider transport for managed-agent calls. These calls can + /// legitimately take longer than the generic governed HTTP timeout while the + /// provider thinks and emits tool use, but they still keep the same public-DNS + /// guard and short connect timeout. + pub fn for_managed_agent() -> Result { + Self::with_timeouts_and_private_networks( + MANAGED_AGENT_REQUEST_TIMEOUT, + DEFAULT_HTTP_CONNECT_TIMEOUT, + false, + None, + ) + } + + /// Build the open-web fetch transport: the optional browser profile (a + /// `Some(user_agent)` enables it; `None` is the plain client) plus the + /// private-network flag. The `http` skill adapter uses this; `new()` and + /// `with_private_network_access()` stay plain for internal/API callers (the + /// agent transport, the registry) where a browser profile does not belong. + pub fn with_options( + allow_private_networks: bool, + browser_user_agent: Option, + ) -> Result { + Self::with_timeouts_and_private_networks( + DEFAULT_HTTP_REQUEST_TIMEOUT, + DEFAULT_HTTP_CONNECT_TIMEOUT, + allow_private_networks, + browser_user_agent, ) } @@ -180,7 +287,7 @@ impl ReqwestHttpTransport { request_timeout: Duration, connect_timeout: Duration, ) -> Result { - Self::with_timeouts_and_private_networks(request_timeout, connect_timeout, true) + Self::with_timeouts_and_private_networks(request_timeout, connect_timeout, true, None) } } @@ -215,7 +322,7 @@ impl RuntimeHttpTransport for ReqwestHttpTransport { .send() .await .map_err(|error| RuntimeHttpError::Transport { - message: error.to_string(), + message: transport_error_message(&error), })?; let status = response.status().as_u16(); let body = read_limited_response_body(response, MAX_HTTP_RESPONSE_BYTES).await?; @@ -224,6 +331,18 @@ impl RuntimeHttpTransport for ReqwestHttpTransport { } } +#[cfg(feature = "async-http")] +fn transport_error_message(error: &(dyn StdError + 'static)) -> String { + let mut parts = vec![error.to_string()]; + let mut source = error.source(); + while let Some(error) = source { + parts.push(error.to_string()); + source = error.source(); + } + parts.dedup(); + parts.join(": ") +} + #[cfg(feature = "async-http")] #[derive(Clone, Debug)] struct GuardedDnsResolver { @@ -1011,4 +1130,87 @@ mod tests { assert!(matches!(error, Some(RuntimeHttpError::Transport { .. }))); Ok(()) } + + #[cfg(feature = "async-http")] + #[test] + fn browser_profile_sends_chrome_ua_and_client_hints() -> Result<(), RuntimeHttpTestError> { + let listener = TcpListener::bind("127.0.0.1:0")?; + let address = listener.local_addr()?; + let server = std::thread::spawn(move || -> Result { + let (mut stream, _) = listener.accept()?; + let mut buffer = [0_u8; 4096]; + let bytes_read = stream.read(&mut buffer)?; + stream.write_all(b"HTTP/1.1 204 No Content\r\nContent-Length: 0\r\n\r\n")?; + Ok(String::from_utf8_lossy(&buffer[..bytes_read]).into_owned()) + }); + + // with_options(private = true) so the loopback test server is reachable. + let transport = ReqwestHttpTransport::with_options( + true, + Some(super::DEFAULT_BROWSER_USER_AGENT.to_owned()), + )?; + transport.send(RuntimeHttpRequest { + method: HttpMethod::Get, + url: format!("http://{address}/probe"), + headers: Vec::new(), + body: None, + })?; + let request = server + .join() + .map_err(|_| RuntimeHttpTestError::ServerThread)??; + + let lower = request.to_ascii_lowercase(); + assert!( + lower.contains("chrome/143"), + "browser UA should be sent: {request}" + ); + assert!( + lower.contains("sec-ch-ua"), + "client-hint headers should be sent: {request}" + ); + assert!( + lower.contains("sec-fetch-mode"), + "fetch-metadata headers should be sent: {request}" + ); + Ok(()) + } + + #[cfg(feature = "async-http")] + #[test] + fn caller_header_overrides_browser_default() -> Result<(), RuntimeHttpTestError> { + let listener = TcpListener::bind("127.0.0.1:0")?; + let address = listener.local_addr()?; + let server = std::thread::spawn(move || -> Result { + let (mut stream, _) = listener.accept()?; + let mut buffer = [0_u8; 4096]; + let bytes_read = stream.read(&mut buffer)?; + stream.write_all(b"HTTP/1.1 204 No Content\r\nContent-Length: 0\r\n\r\n")?; + Ok(String::from_utf8_lossy(&buffer[..bytes_read]).into_owned()) + }); + + let transport = ReqwestHttpTransport::with_options( + true, + Some(super::DEFAULT_BROWSER_USER_AGENT.to_owned()), + )?; + transport.send(RuntimeHttpRequest { + method: HttpMethod::Get, + url: format!("http://{address}/probe"), + headers: vec![RuntimeHttpHeader::new("accept", "application/json")], + body: None, + })?; + let request = server + .join() + .map_err(|_| RuntimeHttpTestError::ServerThread)??; + + let lower = request.to_ascii_lowercase(); + assert!( + lower.contains("accept: application/json"), + "caller Accept should be present: {request}" + ); + assert!( + !lower.contains("text/html"), + "browser default Accept should be overridden, not duplicated: {request}" + ); + Ok(()) + } } diff --git a/crates/runx-runtime/src/scaffold.rs b/crates/runx-runtime/src/scaffold.rs index cb502f1e2..f23da9cd4 100644 --- a/crates/runx-runtime/src/scaffold.rs +++ b/crates/runx-runtime/src/scaffold.rs @@ -7,10 +7,7 @@ pub use init::{ InitAction, InitGeneratedValues, RunxInitOptions, RunxInitResult, RunxInstallState, RunxProjectState, ensure_runx_install_state, ensure_runx_project_state, runx_init, }; -pub use new::{ - RunxNewOptions, RunxNewResult, packet_namespace_for_name, sanitize_runx_package_name, - scaffold_runx_package, -}; +pub use new::{RunxNewOptions, RunxNewResult, sanitize_runx_package_name, scaffold_runx_package}; use std::fmt; use std::io; diff --git a/crates/runx-runtime/src/scaffold/new.rs b/crates/runx-runtime/src/scaffold/new.rs index 0dada3fbb..aeca05429 100644 --- a/crates/runx-runtime/src/scaffold/new.rs +++ b/crates/runx-runtime/src/scaffold/new.rs @@ -4,20 +4,17 @@ use std::path::{Path, PathBuf}; use serde::Serialize; use super::ScaffoldError; -use super::templates::{ScaffoldTemplateVersions, scaffold_package_files}; +use super::templates::scaffold_package_files; #[derive(Clone, Debug, PartialEq, Eq)] pub struct RunxNewOptions { pub name: String, pub directory: PathBuf, - pub authoring_package_version: String, - pub cli_package_version: String, } #[derive(Clone, Debug, PartialEq, Eq, Serialize)] pub struct RunxNewResult { pub name: String, - pub packet_namespace: String, pub directory: PathBuf, pub files: Vec, pub next_steps: Vec, @@ -25,20 +22,10 @@ pub struct RunxNewResult { pub fn scaffold_runx_package(options: &RunxNewOptions) -> Result { let name = sanitize_runx_package_name(&options.name); - let packet_namespace = packet_namespace_for_name(&name); let root = lexical_absolute(&options.directory)?; assert_writable_scaffold_target(&root)?; - let versions = ScaffoldTemplateVersions { - authoring_package_version: options.authoring_package_version.clone(), - authoring_toolkit_version: options - .authoring_package_version - .strip_prefix('^') - .unwrap_or(&options.authoring_package_version) - .to_owned(), - cli_package_version: options.cli_package_version.clone(), - }; - let writes = scaffold_package_files(&name, &packet_namespace, &versions); + let writes = scaffold_package_files(&name); fs::create_dir_all(&root) .map_err(|source| ScaffoldError::io("creating scaffold root", &root, source))?; @@ -48,14 +35,12 @@ pub fn scaffold_runx_package(options: &RunxNewOptions) -> Result String { } } -#[must_use] -pub fn packet_namespace_for_name(value: &str) -> String { - let unscoped = value.to_lowercase().trim_start_matches('@').to_owned(); - let namespace = trim_dots(&replace_runs( - &unscoped, - |character| character.is_ascii_lowercase() || character.is_ascii_digit(), - '.', - )); - if namespace.is_empty() { - "runx.package".to_owned() - } else { - namespace - } -} - fn assert_writable_scaffold_target(root: &Path) -> Result<(), ScaffoldError> { match fs::read_dir(root) { Ok(mut entries) => match entries.next() { @@ -151,7 +121,3 @@ fn trim_boundary_separators(value: &str) -> String { .trim_matches(|character| matches!(character, '.' | '_' | '-')) .to_owned() } - -fn trim_dots(value: &str) -> String { - value.trim_matches('.').to_owned() -} diff --git a/crates/runx-runtime/src/scaffold/templates.rs b/crates/runx-runtime/src/scaffold/templates.rs index a3eb131b9..5bc6bd011 100644 --- a/crates/runx-runtime/src/scaffold/templates.rs +++ b/crates/runx-runtime/src/scaffold/templates.rs @@ -1,14 +1,6 @@ -// rust-style-allow: large-file because the scaffold templates intentionally -// mirror the TypeScript scaffolder's checked output byte-for-byte. - -use sha2::{Digest, Sha256}; - -#[derive(Clone, Debug, PartialEq, Eq)] -pub struct ScaffoldTemplateVersions { - pub authoring_package_version: String, - pub authoring_toolkit_version: String, - pub cli_package_version: String, -} +// Native cli-tool skill scaffold: SKILL.md + X.yaml + run.mjs + README + .gitignore. +// The output has zero dependencies and no build step, so `runx new` produces a +// skill that runs and harnesses immediately, with nothing pinned that can drift. #[derive(Clone, Debug, PartialEq, Eq)] pub struct ScaffoldFile { @@ -16,59 +8,13 @@ pub struct ScaffoldFile { pub contents: String, } -pub fn scaffold_package_files( - name: &str, - packet_namespace: &str, - versions: &ScaffoldTemplateVersions, -) -> Vec { - let packet_id = format!("{packet_namespace}.echo.v1"); - let tool_source = tool_source(&packet_id); - let tool_runtime = tool_runtime(&packet_id); - let source_hash = source_hash(&tool_source, &tool_runtime); - let schema_hash = schema_hash(&packet_id); - let prompt_fingerprint = prompt_fingerprint(&packet_id); +pub fn scaffold_package_files(name: &str) -> Vec { vec![ - file( - "package.json", - package_json( - name, - &versions.authoring_package_version, - &versions.cli_package_version, - ), - ), - file("README.md", readme(name)), file("SKILL.md", skill_md(name)), file("X.yaml", x_yaml(name)), - file("src/packets/echo.ts", packet_source(&packet_id)), - file( - "dist/packets/echo.v1.schema.json", - packet_schema(packet_namespace, &packet_id), - ), - file("tools/docs/echo/src/index.ts", tool_source.clone()), - file("tools/docs/echo/run.mjs", tool_runtime), - file( - "tools/docs/echo/manifest.json", - tool_manifest( - &packet_id, - &source_hash, - &schema_hash, - &versions.authoring_toolkit_version, - ), - ), - file("tools/docs/echo/fixtures/basic.yaml", tool_fixture(&packet_id)), - file("fixtures/agent.yaml", agent_fixture_yaml(&packet_id)), - file( - "fixtures/agent.replay.json", - agent_replay_json(&packet_id, &prompt_fingerprint), - ), - file("fixtures/repos/readme-only/README.md", format!("# {name}\n")), - file(".github/workflows/publish.yml", publish_workflow()), + file("run.mjs", run_mjs()), + file("README.md", readme(name)), file(".gitignore", "node_modules/\n.runx/\n*.tgz\n".to_owned()), - file( - ".gitattributes", - "tools/**/run.mjs linguist-generated=true\ntools/**/manifest.json linguist-generated=true\ntools/**/dist/** linguist-generated=true\n".to_owned(), - ), - file("tsconfig.json", tsconfig_json()), ] } @@ -79,86 +25,37 @@ fn file(relative_path: &str, contents: String) -> ScaffoldFile { } } -fn package_json(name: &str, authoring_version: &str, cli_version: &str) -> String { - format!( - r#"{{ - "name": "{name}", - "version": "0.1.0", - "description": "Scaffolded runx skill package.", - "type": "module", - "publishConfig": {{ - "access": "public" - }}, - "scripts": {{ - "build": "runx tool build --all --json", - "runx:list": "runx list --json", - "runx:doctor": "runx doctor --json", - "runx:dev": "runx dev --lane deterministic --json", - "prepublishOnly": "runx tool build --all --json && runx doctor --json" - }}, - "runx": {{ - "packets": [ - "./dist/packets/*.schema.json" - ] - }}, - "devDependencies": {{ - "@runxhq/authoring": "{authoring_version}", - "@runxhq/cli": "{cli_version}", - "@tsconfig/node20": "^20.1.6", - "tsx": "^4.20.6" - }} -}}"# - ) -} - -fn readme(name: &str) -> String { - format!( - r#"# {name} - -Runx authoring package: composable skills governed by typed contracts. - -## Layout - -- `SKILL.md`: Anthropic-standard skill description. Read by humans and agents. -- `X.yaml`: runx execution profile layered on top of `SKILL.md`. -- `src/packets/`: typed packet contracts authored with TypeBox. -- `tools/`: deterministic implementation units authored with `defineTool`. -- `fixtures/`: examples and tests across deterministic, agent, and repo-integration lanes. - -## Authoring Loop - -```bash -pnpm install -pnpm build -pnpm runx:list -pnpm runx:doctor -pnpm runx:dev -``` - -Edit `tools/docs/echo/src/index.ts`, then run `runx tool build --all` to regenerate `manifest.json` and `run.mjs`. Add fixtures in `tools///fixtures/` to lock behaviour. - -Packet IDs are immutable. Schema changes mean a new packet ID, not an in-place edit. - -## Bootstrap - -- Canonical: `runx new {name}` -- Cold start: `npm create @runxhq/skill@latest {name}` - -## Publish - -The scaffold includes `.github/workflows/publish.yml`, which publishes with npm provenance from GitHub Actions. Before publishing, update `package.json` metadata for your repo and package. -"# - ) -} - fn skill_md(name: &str) -> String { format!( r#"--- name: {name} -description: Scaffolded runx skill package. +description: {name} runx skill. Replace this with what the skill does and returns. +source: + type: cli-tool + command: node + args: + - run.mjs + timeout_seconds: 30 + sandbox: + profile: readonly + cwd_policy: skill-directory +inputs: + message: + type: string + required: true + description: Input the skill acts on. Replace with the real inputs. +runx: + category: ops + input_resolution: + required: + - message --- -Use this skill to demonstrate a governed runx authoring package. +# {name} + +Describe what this skill does, when an agent should reach for it, and what it +returns. Replace the echo in `run.mjs` with the real work, and add cases to +`X.yaml` so the behaviour is locked by the harness. "# ) } @@ -166,296 +63,92 @@ Use this skill to demonstrate a governed runx authoring package. fn x_yaml(name: &str) -> String { format!( r#"skill: {name} +version: "0.1.0" + +catalog: + kind: skill + audience: public + visibility: public + role: canonical + +harness: + cases: + - name: {name}-smoke + runner: default + inputs: + message: hello + expect: + status: sealed + receipt: + schema: runx.receipt.v1 + state: sealed + disposition: closed + reason_code: process_closed + - name: {name}-empty-message-fails + runner: default + inputs: + message: "" + expect: + status: failure + receipt: + schema: runx.receipt.v1 + state: sealed + disposition: closed + reason_code: process_failed runners: default: default: true - type: graph + type: cli-tool + command: node + args: + - run.mjs inputs: message: type: string - required: false - default: hello - graph: - name: {name} - steps: - - id: echo - tool: docs.echo - inputs: - message: inputs.message + required: true + description: Input the skill acts on. "# ) } -fn packet_source(packet_id: &str) -> String { - format!( - r#"import {{ definePacket, t }} from "@runxhq/authoring"; - -export const EchoPacket = definePacket({{ - id: "{packet_id}", - schema: t.Object({{ - message: t.String(), - }}), -}}); -"# - ) -} - -fn packet_schema(packet_namespace: &str, packet_id: &str) -> String { - let schema_path = packet_namespace.replace('.', "/"); - format!( - r#"{{ - "$schema": "https://json-schema.org/draft/2020-12/schema", - "$id": "https://schemas.runx.dev/{schema_path}/echo/v1.json", - "x-runx-packet-id": "{packet_id}", - "type": "object", - "required": [ - "message" - ], - "properties": {{ - "message": {{ - "type": "string" - }} - }}, - "additionalProperties": false -}}"# - ) -} - -fn tool_source(packet_id: &str) -> String { - format!( - r#"import {{ defineTool, stringInput }} from "@runxhq/authoring"; - -export default defineTool({{ - name: "docs.echo", - version: "0.1.0", - description: "Echo a docs message.", - inputs: {{ - message: stringInput({{ default: "hello" }}), - }}, - output: {{ - packet: "{packet_id}", - wrap_as: "echo_packet", - }}, - scopes: ["docs.read"], - run({{ inputs }}) {{ - return {{ message: inputs.message }}; - }}, -}}); -"# - ) +fn run_mjs() -> String { + r#"// Inputs arrive as RUNX_INPUT_ environment variables. Do the work and +// write the result to stdout. Replace this echo with the real logic. +const message = process.env.RUNX_INPUT_MESSAGE ?? ""; +if (message.trim().length === 0) { + process.stderr.write("message is required\n"); + process.exit(64); } - -fn tool_runtime(packet_id: &str) -> String { - format!( - r#"const fs = require("node:fs"); -const rawInputs = process.env.RUNX_INPUTS_PATH - ? fs.readFileSync(process.env.RUNX_INPUTS_PATH, "utf8") - : (process.env.RUNX_INPUTS_JSON || "{{}}"); -const inputs = JSON.parse(rawInputs); -process.stdout.write(JSON.stringify({{ schema: "{packet_id}", data: {{ message: String(inputs.message || "hello") }} }})); +process.stdout.write(`${message}\n`); "# - ) + .to_owned() } -fn tool_manifest( - packet_id: &str, - source_hash: &str, - schema_hash: &str, - toolkit_version: &str, -) -> String { +fn readme(name: &str) -> String { format!( - r#"{{ - "schema": "runx.tool.manifest.v1", - "name": "docs.echo", - "version": "0.1.0", - "description": "Echo a docs message.", - "source": {{ - "type": "cli-tool", - "command": "node", - "args": [ - "./run.mjs" - ] - }}, - "runtime": {{ - "command": "node", - "args": [ - "./run.mjs" - ] - }}, - "inputs": {{ - "message": {{ - "type": "string", - "required": false, - "default": "hello" - }} - }}, - "output": {{ - "packet": "{packet_id}", - "wrap_as": "echo_packet" - }}, - "scopes": [ - "docs.read" - ], - "runx": {{ - "artifacts": {{ - "wrap_as": "echo_packet" - }} - }}, - "source_hash": "{source_hash}", - "schema_hash": "{schema_hash}", - "toolkit_version": "{toolkit_version}" -}}"# - ) -} + r#"# {name} -fn tool_fixture(packet_id: &str) -> String { - format!( - r#"name: echo-basic -lane: deterministic -target: - kind: tool - ref: docs.echo -inputs: - message: hello -expect: - status: sealed - output: - subset: - schema: {packet_id} - data: - message: hello -"# - ) -} +A native runx skill: a `SKILL.md` contract, an `X.yaml` execution profile, and a +`run.mjs` script. No build step and no dependencies. -fn agent_fixture_yaml(packet_id: &str) -> String { - format!( - r#"name: echo-agent-replay -lane: agent -target: - kind: skill - ref: . -inputs: - message: hello -agent: - mode: replay -expect: - status: sealed - outputs: - echo_packet: - matches_packet: {packet_id} -"# - ) -} +## Develop -fn agent_replay_json(packet_id: &str, prompt_fingerprint: &str) -> String { - format!( - r#"{{ - "schema": "runx.replay.v1", - "fixture": "echo-agent-replay", - "prompt_fingerprint": "{prompt_fingerprint}", - "recorded_at": "1970-01-01T00:00:00.000Z", - "target": {{ - "kind": "skill", - "ref": "." - }}, - "status": "sealed", - "outputs": {{ - "echo_packet": {{ - "schema": "{packet_id}", - "data": {{ - "message": "hello" - }} - }} - }}, - "usage": {{ - "mode": "scaffold" - }} -}}"# - ) -} +```bash +runx harness . --json # run the harness cases in X.yaml +runx skill . --input message=hello --json # run the skill once +runx history # inspect the signed receipt +``` -fn publish_workflow() -> String { - r#"name: publish +Edit `run.mjs` to do the real work, and keep both harness classes in `X.yaml`: +one happy path and one stop, error, or refusal case. -on: - workflow_dispatch: - release: - types: - - published +## Publish -jobs: - publish: - runs-on: ubuntu-latest - permissions: - contents: read - id-token: write - steps: - - uses: actions/checkout@v4 - - uses: pnpm/action-setup@v4 - with: - version: 10 - - uses: actions/setup-node@v4 - with: - node-version: 20 - registry-url: https://registry.npmjs.org - cache: pnpm - - run: pnpm install --frozen-lockfile - - run: pnpm build - - run: pnpm runx:doctor - - run: npm publish --provenance --access public - env: - NODE_AUTH_TOKEN: ${{ secrets.NPM_TOKEN }} +```bash +runx login --provider github --for publish +runx registry publish . # the registry runs the harness as the publish gate +``` "# - .to_owned() -} - -fn tsconfig_json() -> String { - r#"{ - "extends": "@tsconfig/node20/tsconfig.json", - "compilerOptions": { - "module": "NodeNext", - "moduleResolution": "NodeNext", - "strict": true - }, - "include": [ - "src/**/*.ts", - "tools/**/*.ts" - ] -}"# - .to_owned() -} - -fn source_hash(tool_source: &str, tool_runtime: &str) -> String { - let mut hasher = Sha256::new(); - hasher.update("src/index.ts"); - hasher.update([0]); - hasher.update(tool_source); - hasher.update([0]); - hasher.update("run.mjs"); - hasher.update([0]); - hasher.update(tool_runtime); - hasher.update([0]); - format!("sha256:{:x}", hasher.finalize()) -} - -fn schema_hash(packet_id: &str) -> String { - let stable = format!( - r#"{{"artifacts":{{"wrap_as":"echo_packet"}},"inputs":{{"message":{{"default":"hello","required":false,"type":"string"}}}},"output":{{"packet":"{packet_id}","wrap_as":"echo_packet"}}}}"# - ); - format!("sha256:{}", hash_string(&stable)) -} - -fn prompt_fingerprint(packet_id: &str) -> String { - let stable = format!( - r#"{{"agent":{{"mode":"replay"}},"expect":{{"outputs":{{"echo_packet":{{"matches_packet":"{packet_id}"}}}},"status":"sealed"}},"inputs":{{"message":"hello"}},"target":{{"kind":"skill","ref":"."}}}}"# - ); - format!("sha256:{}", hash_string(&stable)) -} - -fn hash_string(value: &str) -> String { - let mut hasher = Sha256::new(); - hasher.update(value); - format!("{:x}", hasher.finalize()) + ) } diff --git a/crates/runx-runtime/src/tool_catalogs/build.rs b/crates/runx-runtime/src/tool_catalogs/build.rs index a224c950a..425d8631f 100644 --- a/crates/runx-runtime/src/tool_catalogs/build.rs +++ b/crates/runx-runtime/src/tool_catalogs/build.rs @@ -86,8 +86,6 @@ fn build_tool_manifest( let manifest_path = tool_dir.join("manifest.json"); let source = fs::read_to_string(&manifest_path) .map_err(|error| ToolCatalogError::io("reading tool manifest", &manifest_path, error))?; - let raw: RawToolManifest = serde_json::from_str(&source) - .map_err(|error| ToolCatalogError::json("parsing tool manifest", &manifest_path, error))?; let raw_payload: JsonPayload = serde_json::from_str(&source) .map_err(|error| ToolCatalogError::json("parsing tool manifest", &manifest_path, error))?; let JsonPayload::Object(raw_object) = raw_payload else { @@ -96,6 +94,13 @@ fn build_tool_manifest( message: "manifest.json must be an object.".to_owned(), }); }; + let normalized_object = normalize_tool_manifest_shape(raw_object.clone()); + let raw: RawToolManifest = serde_json::from_value( + serde_json::to_value(JsonPayload::Object(normalized_object.clone())).map_err(|error| { + ToolCatalogError::json("normalizing tool manifest", &manifest_path, error) + })?, + ) + .map_err(|error| ToolCatalogError::json("parsing tool manifest", &manifest_path, error))?; let output = raw .output .unwrap_or_else(|| normalize_tool_output(raw.runx.as_ref())); @@ -132,6 +137,29 @@ fn build_tool_manifest( }) } +fn normalize_tool_manifest_shape(mut raw: JsonPayloadObject) -> JsonPayloadObject { + let Some(JsonPayload::Object(source)) = raw.get_mut("source") else { + return raw; + }; + if !matches!( + source.get("type"), + Some(JsonPayload::String(value)) if value == "http" + ) || source.contains_key("http") + { + return raw; + } + let mut http = JsonPayloadObject::new(); + for key in ["url", "method", "headers", "allow_private_network"] { + if let Some(value) = source.remove(key) { + http.insert(key.to_owned(), value); + } + } + if !http.is_empty() { + source.insert("http".to_owned(), JsonPayload::Object(http)); + } + raw +} + fn normalize_tool_output(runx: Option<&JsonPayloadObject>) -> ToolOutput { let artifacts = runx .and_then(|runx| runx.get("artifacts")) diff --git a/crates/runx-runtime/tests/local_credential_provision.rs b/crates/runx-runtime/tests/local_credential_provision.rs index 05a731b3e..eb64f9e1a 100644 --- a/crates/runx-runtime/tests/local_credential_provision.rs +++ b/crates/runx-runtime/tests/local_credential_provision.rs @@ -178,7 +178,7 @@ fn graph_http_credential_does_not_break_local_cli_tool_steps() )] .into_iter() .collect(), - env: http_private_network_grant_env(), + env: mixed_http_cli_graph_env(), cwd: temp.path().to_path_buf(), local_credential: Some(LocalCredentialDescriptor { provider: "example-crm".to_owned(), @@ -226,6 +226,13 @@ fn local_sandbox_fallback_env() -> BTreeMap { .into() } +#[cfg(feature = "http")] +fn mixed_http_cli_graph_env() -> BTreeMap { + let mut env = http_private_network_grant_env(); + env.extend(local_sandbox_fallback_env()); + env +} + /// A cli-tool skill that echoes the delivered `$GITHUB_TOKEN`. The command is a /// local shell process: no network, no hosted dependency. fn write_echo_token_skill(root: &Path) -> Result> { diff --git a/crates/runx-runtime/tests/registry.rs b/crates/runx-runtime/tests/registry.rs index 33297f61c..91d45fa92 100644 --- a/crates/runx-runtime/tests/registry.rs +++ b/crates/runx-runtime/tests/registry.rs @@ -105,6 +105,8 @@ fn file_registry_store_covers_profiled_skill_surface() -> Result<(), Box Result<(), Box Result<(), Box Result<(), Box Result<(), Box, next_steps: Vec, } #[test] -fn new_scaffold_matches_typescript_fixture() -> Result<(), Box> { +fn new_scaffold_matches_native_fixture() -> Result<(), Box> { let temp = TestDir::create("new-byte-parity")?; let target = temp.path().join("docs-demo"); let options = RunxNewOptions { name: "docs-demo".to_owned(), directory: target.clone(), - authoring_package_version: "^0.1.4".to_owned(), - cli_package_version: "^0.5.22".to_owned(), }; let result = scaffold_runx_package(&options)?; let manifest = scaffold_fixture_manifest()?; assert_eq!(result.name, manifest.name); - assert_eq!(result.packet_namespace, manifest.packet_namespace); assert_eq!(result.files, manifest.files); assert_eq!( normalize_next_steps(&target, &result.next_steps), @@ -53,8 +49,6 @@ fn new_refuses_non_empty_targets() -> Result<(), Box> { let options = RunxNewOptions { name: "docs-demo".to_owned(), directory: target.clone(), - authoring_package_version: "^0.1.4".to_owned(), - cli_package_version: "^0.5.22".to_owned(), }; match scaffold_runx_package(&options) { @@ -62,7 +56,7 @@ fn new_refuses_non_empty_targets() -> Result<(), Box> { Err(error) => return Err(format!("expected non-empty target error, got {error}").into()), Ok(_) => return Err("expected non-empty target error".into()), } - assert!(!target.join("package.json").exists()); + assert!(!target.join("SKILL.md").exists()); Ok(()) } diff --git a/crates/runx-runtime/tests/skill_run.rs b/crates/runx-runtime/tests/skill_run.rs index b2411d1b4..11efcad36 100644 --- a/crates/runx-runtime/tests/skill_run.rs +++ b/crates/runx-runtime/tests/skill_run.rs @@ -392,6 +392,9 @@ fn native_skill_run_treats_structured_stdout_as_claim_not_receipt_proof() "title": "Injected source" } ] + }, + "closure": { + "disposition": "closed" } } } @@ -499,6 +502,9 @@ fn native_skill_run_uses_runtime_receipt_path_resolution() -> Result<(), Box Result<(), Box Result<(), Box Result<(), Box Result SkillRunRequest { crate::support::insert_test_signing_env(&mut request.env); request + .env + .entry("RUNX_HOME".to_owned()) + .or_insert_with(|| request.cwd.join(".runx").to_string_lossy().into_owned()); + request } fn write_agent_task_skill(root: &Path) -> Result> { @@ -3022,8 +3059,18 @@ fn native_graph_when_skips_unselected_branch() -> Result<(), Box Result<(), Box> { - let temp_root = copy_scaffold_fixture("build_scaffold_manifest")?; - let tool_dir = temp_root.join("tools/docs/echo"); +fn tool_catalogs_build_minimal_manifest() -> Result<(), Box> { + let temp_root = copy_minimal_tool_fixture("build_minimal_manifest")?; + let tool_dir = temp_root.join("tools/fixture/minimal"); let report = build_tool_catalogs(&ToolBuildOptions { root: temp_root.clone(), @@ -23,11 +23,11 @@ fn tool_catalogs_build_scaffold_manifest() -> Result<(), Box Result<(), Box Result<(), Box> { - let temp_root = copy_scaffold_fixture("inspect_local_manifest")?; + let temp_root = copy_minimal_tool_fixture("inspect_local_manifest")?; let report = inspect_tool(&ToolInspectOptions { root: temp_root.clone(), - tool_ref: "docs.echo".to_owned(), + tool_ref: "fixture.minimal".to_owned(), source: None, search_from_directory: temp_root.clone(), tool_roots: Vec::new(), @@ -89,11 +89,11 @@ fn tool_catalogs_inspect_local_manifest() -> Result<(), Box Result<(), Box> { - let temp_root = copy_scaffold_fixture("reject_absolute_manifest_path")?; - let manifest = temp_root.join("tools/docs/echo/manifest.json"); + let temp_root = copy_minimal_tool_fixture("reject_absolute_manifest_path")?; + let manifest = temp_root.join("tools/fixture/minimal/manifest.json"); let error = match inspect_tool(&ToolInspectOptions { root: temp_root.clone(), @@ -181,7 +181,7 @@ fn tool_catalogs_reject_absolute_explicit_manifest_path() -> Result<(), Box Result<(), Box> { - let temp_root = copy_scaffold_fixture("reject_parent_manifest_path")?; + let temp_root = copy_minimal_tool_fixture("reject_parent_manifest_path")?; let error = match inspect_tool(&ToolInspectOptions { root: temp_root.clone(), @@ -208,7 +208,7 @@ fn tool_catalogs_reject_parent_traversal_explicit_manifest_path() #[test] fn tool_catalogs_inspect_prefers_local_manifest_over_fixture_catalog() -> Result<(), Box> { - let temp_root = copy_scaffold_fixture("inspect_local_precedence")?; + let temp_root = copy_minimal_tool_fixture("inspect_local_precedence")?; let tool_dir = temp_root.join("tools/fixture/echo"); fs::create_dir_all(&tool_dir)?; fs::write( @@ -261,8 +261,8 @@ fn tool_catalogs_inspect_prefers_local_manifest_over_fixture_catalog() Ok(()) } -fn copy_scaffold_fixture(name: &str) -> Result> { - let source = repo_root()?.join("fixtures/scaffold/new-docs-demo/files"); +fn copy_minimal_tool_fixture(name: &str) -> Result> { + let source = repo_root()?.join("fixtures/tool-catalogs/build/minimal/workspace"); let target = std::env::temp_dir() .join("runx-tool-catalogs-tests") .join(format!("{name}-{}", std::process::id())); diff --git a/docs/api-surface.md b/docs/api-surface.md index 5486adb8a..7afd1f9fc 100644 --- a/docs/api-surface.md +++ b/docs/api-surface.md @@ -34,16 +34,6 @@ Version: `0.3.0` | --- | --- | --- | | `@runxhq/contracts` | `./dist/index.d.ts` | `./dist/index.js` | -## @runxhq/create-skill - -Cold-start scaffolder for runx standalone skill packages. - -Version: `0.2.0` - -| Import | Types | Runtime | -| --- | --- | --- | -| `@runxhq/create-skill` | `./dist/index.d.ts` | `./dist/index.js` | - ## @runxhq/host-adapters Thin host response adapters over the runx host protocol. diff --git a/docs/assets/ops-fanout.svg b/docs/assets/ops-fanout.svg new file mode 100644 index 000000000..5c178b325 --- /dev/null +++ b/docs/assets/ops-fanout.svg @@ -0,0 +1,79 @@ + + runx business operator graph + A clean graph showing a business goal entering the runx business-ops skill, fanning into governed lanes, and resolving into receipt or approval outcomes. + + + + + + + + + + + goal + prepare API v2 + + + + + + runx skill + business-ops + intake, plan, route, gate + + + + + + + + + + + + + docs from evidence + + + release readiness + + + receipt audit + + + issue-to-PR + + + customer comms + + + spend review + + + + + + receipt + prove the run + + + approval + approve first + diff --git a/docs/demo-inventory.json b/docs/demo-inventory.json index b51373226..51092145f 100644 --- a/docs/demo-inventory.json +++ b/docs/demo-inventory.json @@ -6,6 +6,11 @@ "proof": "Native CLI top-level skill and harness baseline.", "command": "runx harness examples/hello-world" }, + { + "path": "skills/business-ops", + "proof": "One business signal fans out through governed ops lanes and seals a graph receipt.", + "command": "runx harness skills/business-ops" + }, { "path": "examples/github-mcp-hero", "proof": "Governed GitHub MCP read succeeds, out-of-scope write is refused, denial receipt verifies offline.", diff --git a/docs/demos.md b/docs/demos.md index b9bb65cec..983539206 100644 --- a/docs/demos.md +++ b/docs/demos.md @@ -4,9 +4,8 @@ These demos are runnable from this repository and produce signed receipts. Use t standalone verifier at `tools/verify/verify.mjs` with the demo issuer key in `tools/verify/runx-demo-jwks.json`. -`docs/demo-inventory.json` is the machine-checked source of truth for which -`examples/*` directories are featured demos, runnable previews, or fixture -support. +`docs/demo-inventory.json` is the machine-checked source of truth for featured +demos, runnable previews, and fixture support. ```sh export RUNX_RECEIPT_SIGN_KID=runx-demo-key @@ -19,6 +18,7 @@ export RUNX_RECEIPT_SIGN_ISSUER_TYPE=hosted | Demo | Proof | Run | Gate | | --- | --- | --- | --- | | `examples/hello-world` | Native CLI top-level skill and harness baseline. | `runx harness examples/hello-world` | harness | +| `skills/business-ops` | One business signal fans out through governed ops lanes and seals a graph receipt. | `runx harness skills/business-ops` | harness | | `examples/github-mcp-hero` | GitHub MCP repo read succeeds, out-of-scope write is refused, and the denial receipt verifies offline. | `sh examples/github-mcp-hero/run.sh` | harness | | `examples/http-graph` | A graph step uses the governed HTTP front against a local fixture and seals a receipt tree. | `sh examples/http-graph/run.sh` | harness | | `examples/openapi-graph` | An OpenAPI-described operation is executed through the governed external-adapter lane and sealed. | `sh examples/openapi-graph/run.sh` | harness | diff --git a/docs/getting-started.md b/docs/getting-started.md index 87e7dadb6..ec11ec521 100644 --- a/docs/getting-started.md +++ b/docs/getting-started.md @@ -77,11 +77,13 @@ crates/target/debug/runx history --json ## Next -- Use `crates/target/debug/runx new docs-demo` for local standalone skill - scaffolding. -- Use `npm create @runxhq/skill@latest docs-demo` when starting from npm. +- Use `crates/target/debug/runx new docs-demo` to scaffold a native cli-tool + skill (SKILL.md + X.yaml + run.mjs, zero npm deps). To cold-start without + installing runx first, run `npx @runxhq/cli new docs-demo`; it downloads the + launcher and runs the same native scaffold. - Compose the example into a graph with [Skill To Graph](./skill-to-graph.md). - Publish a ready skill from a public repo at https://runx.ai/x/publish, or run - `crates/target/debug/runx login` followed by `crates/target/debug/runx registry publish ... --registry https://runx.ai`. + `crates/target/debug/runx login --for publish` followed by + `crates/target/debug/runx registry publish ... --registry https://api.runx.ai`. See [Publishing](./publishing.md) for the full local and hosted paths. - See [API Surface](./api-surface.md) for public package exports. diff --git a/docs/publishing.md b/docs/publishing.md index 8d5f8a570..4238b2fe3 100644 --- a/docs/publishing.md +++ b/docs/publishing.md @@ -17,14 +17,27 @@ they behave differently. The skill must be real and runnable: -- A valid `SKILL.md` (frontmatter `name`, `description`, and `source`) and an - `X.yaml` execution profile when the skill has a runnable path. Format: +- A valid `SKILL.md` (frontmatter `name`, `description`, and `source`, with + optional portable `category`) and an execution profile (`X.yaml`) when the + skill has a runnable path. Format: https://runx.ai/SKILL.md + Use top-level `category` only for the upstream skill's own portable taxonomy. + Use `runx.category` for the runx catalog category/facet; runx categories are + intentionally specific and should not be inferred from arbitrary upstream + category values. - It passes the harness: ```bash runx harness ./skills/ --json ``` +`X.yaml` is the canonical v1 filename, but the artifact is an execution profile: +runner wiring, typed inputs and outputs, tool/context refs, authority and receipt +mapping, side-effect posture, and harness cases. Do not use it as a strategy +document, target registry, copy deck, generated state file, or secret container. +Keep profile YAML explicit: anchors, aliases, merge keys, custom tags, +multi-document markers, duplicate mapping keys, and unknown profile fields are +not supported. + ## Publish locally first ```bash @@ -56,8 +69,8 @@ For humans, start at https://runx.ai/x/publish. There are two publish lanes: The CLI form keeps the public API token out of command lines: ```bash -runx login -runx registry publish ./skills//SKILL.md --registry https://runx.ai +runx login --for publish +runx registry publish ./skills//SKILL.md --registry https://api.runx.ai ``` For remote publishes the CLI sends a bounded skill package: @@ -85,11 +98,13 @@ instead of trusting a client-supplied summary, while keeping local credentials, fixtures, source trees, and build trash out of the registry. The local harness still runs first for fast feedback. -`runx login` opens the hosted sign-in flow and stores the returned public API -token in the encrypted local config at `public.api_token`. Hosted CLI commands -use token precedence in this order: an explicit `--token` when the command has -one, then `RUNX_PUBLIC_API_TOKEN`, then the stored token from `runx login`. -`runx registry publish` uses the env or stored-token sources. +`runx login --for publish` opens the hosted sign-in flow and stores a +purpose-scoped public API token in the encrypted local config at +`public.api_token`. The token can publish and report skills, but it cannot move +money, mutate hosted billing state, or operate unrelated hosted surfaces. Hosted +CLI commands use token precedence in this order: an explicit `--token` when the +command has one, then `RUNX_PUBLIC_API_TOKEN`, then the stored token from +`runx login`. `runx registry publish` uses the env or stored-token sources. After a public URL publish, use the claim flow from the registry listing to prove control of the source repo and move matching versions toward verified discovery. @@ -103,7 +118,9 @@ runx treats it like every other governed action, with no special-casing: - The **connected identity** proves the publisher namespace. Hosted runx derives the owner from that identity; the request body cannot spoof it. - The **public API token is stored encrypted locally** and masked by `runx config`. - You can also use `RUNX_PUBLIC_API_TOKEN` for CI. + Use `runx login --for publish` for human publishing, or + `RUNX_PUBLIC_API_TOKEN` for CI when you intentionally inject the same narrow + credential. - New publishes start as **community**. Verification and evidence promote discovery; publisher declaration alone never does. - Hosted publishing is rate-limited per publisher identity. A noisy publisher diff --git a/docs/reference.md b/docs/reference.md index 13f695835..a13f14181 100644 --- a/docs/reference.md +++ b/docs/reference.md @@ -99,7 +99,6 @@ Recommended flows: runx init runx init -g --prefetch official runx new docs-demo -npm create @runxhq/skill@latest docs-demo runx list skills runx registry search sourcey --json runx skill sourcey/sourcey@1.0.0 --registry https://runx.example.test --project . --json @@ -145,9 +144,9 @@ contract layer: Rust-owned schema artifacts. - `@runxhq/cli`: npm distribution wrapper and client presentation around the native CLI. -- `@runxhq/authoring`, `@runxhq/create-skill`, `@runxhq/host-adapters`, and - `@runxhq/langchain`: authoring, scaffolding, host presentation, and bridge - packages over language-neutral contracts. +- `@runxhq/authoring`, `@runxhq/host-adapters`, and `@runxhq/langchain`: + authoring, host presentation, and bridge packages over language-neutral + contracts. For the generated package export index, see [docs/api-surface.md](docs/api-surface.md). @@ -220,22 +219,19 @@ publish policy. See [docs/issue-to-pr.md](docs/issue-to-pr.md). ## Standalone Skill Packages -`runx new ` is the canonical standalone package scaffold: +`runx new ` scaffolds a native cli-tool skill (SKILL.md + X.yaml + +run.mjs, zero npm deps, no build step): ```bash runx new docs-demo ``` -For cold-start adoption, the package entrypoint is: +To cold-start without installing runx first, run `npx @runxhq/cli new docs-demo`; +it downloads the launcher and runs the same native scaffold. -```bash -npm create @runxhq/skill@latest docs-demo -``` - -Both entrypoints go through the same scaffolder. Community skills should be -authored and published as standalone packages created this way. The main `runx` -repo is the first-party lane for official skills and runtime code, not the -community package catalog. +Community skills should be authored and published as standalone packages created +this way. The main `runx` repo is the first-party lane for official skills and +runtime code, not the community package catalog. Registry search and install now normalize public trust into three tiers: `first_party`, `verified`, and `community`. Richer provenance and attestation @@ -258,6 +254,11 @@ skills/sourcey/ Direct execution accepts the package directory or `SKILL.md` inside it. Flat `foo.md` skill files are no longer a supported execution surface. +Execution profiles use a strict YAML subset: no anchors, aliases, merge keys, +custom tags, multi-document markers, duplicate mapping keys, or unknown profile +fields. Keep capability and receipt mappings explicit in the runner that uses +them. + See `../docs/skill-profile-model.md` for resolution rules, publication modes, trust tiers, MCP export, and composite skill behavior. See `../docs/evolution-model.md` for the evolve lane, the skill/tool boundary, @@ -380,7 +381,7 @@ runx publish ./.runx/receipts/.json `runx publish` posts the full sealed receipt to `POST /v1/receipts/notarize` with `publish: true`, then prints the public `/r` link and content hash returned by the notary. Configure the hosted API with `RUNX_PUBLIC_API_BASE_URL` (default -`https://runx.ai`) and authenticate with `RUNX_PUBLIC_API_TOKEN` or `--token` +`https://api.runx.ai`) and authenticate with `RUNX_PUBLIC_API_TOKEN` or `--token` (or run `runx login`). For local hosted dogfood only, point at a loopback API and opt into the private diff --git a/docs/releasing.md b/docs/releasing.md index ba148c94b..9991e0783 100644 --- a/docs/releasing.md +++ b/docs/releasing.md @@ -8,12 +8,16 @@ The CLI ships from `github.com/runxhq/runx`. Release tags are `cli-vX.Y.Z` (prefixed so they do not collide with the repo's other release trains). The git tag is the single source of truth for the version. -The same product version is used on every channel: +The same product version is used on every active channel. The release workflow +is secret-gated, so package-manager channels that are not configured are skipped +with a warning instead of blocking the npm/GitHub release. - GitHub Release: `cli-vX.Y.Z` (the hub; serves the raw per-target archives) - npm: `@runxhq/cli@X.Y.Z` (+ `@runxhq/cli-@X.Y.Z`) -- crates.io: `runx-cli X.Y.Z` (`cargo install runx-cli`) -- Homebrew, Scoop, winget, AUR, Docker (GHCR): `X.Y.Z` +- Homebrew, Scoop, winget, AUR, Docker (GHCR): `X.Y.Z` when their channel + credentials are configured +- crates.io: `runx-cli X.Y.Z` (`cargo install runx-cli`) when the crate channel + is configured `runx --version` reports `CARGO_PKG_VERSION`, so the crate and npm versions are stamped from the tag at build time and the number is truthful regardless of how diff --git a/docs/skill-quality-standard.md b/docs/skill-quality-standard.md index 319e4c0f9..e71f71974 100644 --- a/docs/skill-quality-standard.md +++ b/docs/skill-quality-standard.md @@ -70,3 +70,30 @@ The public catalog test enforces the required sections for every skill with `catalog.visibility: public`. Runnable internals belong in owner-local graph stages at `skills//graph//X.yaml`; they are not hidden catalog skills and should not carry public-skill documentation requirements. + +## Execution Profile Discipline + +Use the term **execution profile** for `X.yaml`. The filename stays `X.yaml` for +v1, but public docs and reviews should describe what it is instead of treating +the letter as the concept. + +`X.yaml` owns capability and governance: + +- named runners and default runner choice; +- typed runner inputs and outputs; +- model-vs-deterministic step boundaries; +- tool, adapter, context-skill, and graph wiring; +- authority, approval, and receipt-act mappings; +- side-effect posture: read, draft, plan, mutate, send, pay, or manual-gated; +- inline `harness.cases`. + +Author `X.yaml` in the strict profile YAML subset: no anchors, aliases, merge +keys, custom tags, multi-document markers, duplicate mapping keys, or unknown +profile fields. Capability mappings should be explicit at the runner that uses +them. + +`X.yaml` must not become the home for long strategy, target registries, campaign +copy, generated state, or secrets. Put operating guidance in `SKILL.md`, +`context/`, or `references/`; put deterministic implementation in `tools/` or +explicit runner files. Doctor and catalog review should treat a bloated, +strategy-heavy profile as a maintainability defect even when it parses. diff --git a/docs/skill-to-graph.md b/docs/skill-to-graph.md index c0eb8fc84..0d091480a 100644 --- a/docs/skill-to-graph.md +++ b/docs/skill-to-graph.md @@ -214,3 +214,40 @@ runx skill \ `--secret-env NAME` names an environment variable to deliver as the secret; `--credential`'s final segment is the scope, which must match the tool's declared `scopes`. See `examples/byo-http-tool` and `examples/http-tool-catalog`. + +For repeated local operator runs, keep the secret in project env and put only the +non-secret descriptor in `.runx/credentials.json`: + +```json +{ + "profiles": { + "operator": { + "credential": "frantic:bearer:local://frantic/internal:frantic.review", + "secret_env": "INTERNAL_SYNC_SECRET" + } + } +} +``` + +Then run with `-p operator` (or `--profile operator`). If +`RUNX_CREDENTIAL_PROFILES` is set, runx reads that JSON file instead; otherwise +it checks the project `.runx/credentials.json` and then the global runx home. +The profile file never contains the secret value. + +Use `inputs` for literals, `$input.*` values, and static configuration. Use +`context` when a step needs an earlier step's output: + +```yaml +steps: + - id: select + run: + type: agent-task + - id: review + run: + type: agent-task + context: + bounty: select.result +``` + +`inputs: { bounty: select.result }` is rejected because it looks like a step +output reference placed in the wrong field. diff --git a/docs/thread-story-contract.md b/docs/thread-story-contract.md index 5ede8a027..0b005a1f1 100644 --- a/docs/thread-story-contract.md +++ b/docs/thread-story-contract.md @@ -47,14 +47,15 @@ TypeScript helper package. Frantic uses that provider lane for source-thread continuity. Frantic emits a typed outbox (`thread.create`, `thread.comment`, `thread.labels`, -`thread.close`) derived from its ledger; runx maps each intent to a provider -push frame with `tools/thread/frantic_thread_outbox.mjs`. `thread.create` -creates or updates the missing GitHub issue by deterministic outbox marker and -returns the observed provider locator. Bound lifecycle intents hydrate the -GitHub issue before writing, then apply comments, labels, or completion closure -through the GitHub provider adapter. Frantic remains the completion authority: -a GitHub issue may close only after a Frantic `thread.close` intent, and GitHub -state never completes a Frantic bounty. +`thread.open`, `thread.close`) derived from its ledger; runx maps each intent to +a provider push frame with `tools/thread/frantic_thread_outbox.mjs`. +`thread.create` creates or updates the missing GitHub issue by deterministic +outbox marker and returns the observed provider locator. Bound lifecycle intents +hydrate the GitHub issue before writing, then apply comments, labels, reopening, +or non-claimable closure through the GitHub provider adapter. Frantic remains +the completion authority: a GitHub issue may close only after a Frantic +`thread.close` intent, may reopen only after a Frantic `thread.open` intent, and +GitHub state never completes or reopens a Frantic bounty. The operational driver for that integration is `scripts/frantic-github-thread-sync.mjs` (`pnpm frantic:github-thread-sync` in diff --git a/docs/ts-interop-boundary.md b/docs/ts-interop-boundary.md index f8a568295..5ef53e345 100644 --- a/docs/ts-interop-boundary.md +++ b/docs/ts-interop-boundary.md @@ -115,10 +115,9 @@ truth, so it is recorded here rather than implied. | Package | Disposition | | --- | --- | | `@runxhq/authoring` | Stays as authoring tooling for skills, manifests, protocol fixtures, and generated artifacts until the authoring DX plan decides whether any piece moves to Rust or scafld. It does not own trusted local execution. | -| `@runxhq/cli` | Stays as a platform-aware npm launcher that resolves and execs the Rust binary. It must remain useful from an installed package without TypeScript sources and must fail closed instead of falling back to TypeScript local execution. | +| `@runxhq/cli` | Stays as a platform-aware npm launcher that resolves and execs the Rust binary. It must remain useful from an installed package without TypeScript sources and must fail closed instead of falling back to TypeScript local execution. It also carries the drift-free cold-start: `npx @runxhq/cli new ` downloads the launcher and runs the same native `runx new` scaffold without a prior runx install. | | `@runxhq/contracts` | Stays as the published generated TypeScript view of `runx-contracts`, maintained with fixture cross-validation. | | `@runxhq/core` | Deleted. Its registry/config/parser remnants were not a shipped execution boundary; live OSS code uses Rust crates, generated contracts, tool-local modules, or explicit protocol packages instead. Cloud imports the promoted `@runx/protocol` package. | -| `@runxhq/create-skill` | Stays as a thin npm bootstrapper that wraps `runx new` through the CLI. | | `@runxhq/host-adapters` | Stays as thin host response adapters over the runx host protocol, retargeted to `@runxhq/contracts` types. It can shape host/client responses, not execute trusted local runtime behavior. | | `@runxhq/langchain` | Stays as an optional LangChain bridge that shells the `runx` CLI or uses documented external protocols for governed skill and tool invocation. | | `runx-py` | Stays as a thin Python client over `runx` CLI JSON output. | diff --git a/fixtures/contracts/operational-proposal/invalid-provider-locked-reference-type.json b/fixtures/contracts/operational-proposal/invalid-provider-locked-reference-type.json deleted file mode 100644 index e7c41c051..000000000 --- a/fixtures/contracts/operational-proposal/invalid-provider-locked-reference-type.json +++ /dev/null @@ -1 +0,0 @@ -{"description":"Invalid operational proposal fixture with a provider-locked reference type.","expected":{"authority":{"final_decision_authority_granted":false,"mutation_authority_granted":false,"proposal_only":true,"publication_authority_granted":false},"confidence":0.8,"decision_summary":"Provider-locked reference types must fail inside operational proposals.","hydrated_context_ref":{"type":"artifact","uri":"runx:artifact:hydrated_context_invalid"},"idempotency":{"fingerprint":"sha256:invalid-provider-locked-reference-type","key":"operational-proposal:invalid:provider-locked-reference-type"},"owner_route_id":"api-owner","proposal_id":"proposal_invalid_provider_locked_reference_type","proposal_kind":"escalation","public_summary":"Invalid proposal with a provider-locked reference type.","rationale":"Operational proposal refs use reusable provider-neutral types and carry concrete providers in provider, locator, or uri fields.","recommended_actions":[{"action_intent":"tracking-to-change","mutating":true,"summary":"Build a fix."}],"redaction_status":"redacted","schema":"runx.operational_proposal.v1","source_event_id":"source_event_invalid","source_ref":{"locator":"team/channel/thread","provider":"slack","type":"slack_thread","uri":"slack://team/T123/channel/CBUG/thread/1710000000.000700"}},"fixture_kind":"operational_proposal_invalid","name":"invalid-provider-locked-reference-type","scope":"operational-proposal"} diff --git a/fixtures/scaffold/new-docs-demo/files/.gitattributes b/fixtures/scaffold/new-docs-demo/files/.gitattributes deleted file mode 100644 index 5f1233cc6..000000000 --- a/fixtures/scaffold/new-docs-demo/files/.gitattributes +++ /dev/null @@ -1,3 +0,0 @@ -tools/**/run.mjs linguist-generated=true -tools/**/manifest.json linguist-generated=true -tools/**/dist/** linguist-generated=true diff --git a/fixtures/scaffold/new-docs-demo/files/.github/workflows/publish.yml b/fixtures/scaffold/new-docs-demo/files/.github/workflows/publish.yml deleted file mode 100644 index 3e9c3be3b..000000000 --- a/fixtures/scaffold/new-docs-demo/files/.github/workflows/publish.yml +++ /dev/null @@ -1,30 +0,0 @@ -name: publish - -on: - workflow_dispatch: - release: - types: - - published - -jobs: - publish: - runs-on: ubuntu-latest - permissions: - contents: read - id-token: write - steps: - - uses: actions/checkout@v4 - - uses: pnpm/action-setup@v4 - with: - version: 10 - - uses: actions/setup-node@v4 - with: - node-version: 20 - registry-url: https://registry.npmjs.org - cache: pnpm - - run: pnpm install --frozen-lockfile - - run: pnpm build - - run: pnpm runx:doctor - - run: npm publish --provenance --access public - env: - NODE_AUTH_TOKEN: ${{ secrets.NPM_TOKEN }} diff --git a/fixtures/scaffold/new-docs-demo/files/README.md b/fixtures/scaffold/new-docs-demo/files/README.md index 431b38b84..8b71dcc39 100644 --- a/fixtures/scaffold/new-docs-demo/files/README.md +++ b/fixtures/scaffold/new-docs-demo/files/README.md @@ -1,34 +1,22 @@ # docs-demo -Runx authoring package: composable skills governed by typed contracts. +A native runx skill: a `SKILL.md` contract, an `X.yaml` execution profile, and a +`run.mjs` script. No build step and no dependencies. -## Layout - -- `SKILL.md`: Anthropic-standard skill description. Read by humans and agents. -- `X.yaml`: runx execution profile layered on top of `SKILL.md`. -- `src/packets/`: typed packet contracts authored with TypeBox. -- `tools/`: deterministic implementation units authored with `defineTool`. -- `fixtures/`: examples and tests across deterministic, agent, and repo-integration lanes. - -## Authoring Loop +## Develop ```bash -pnpm install -pnpm build -pnpm runx:list -pnpm runx:doctor -pnpm runx:dev +runx harness . --json # run the harness cases in X.yaml +runx skill . --input message=hello --json # run the skill once +runx history # inspect the signed receipt ``` -Edit `tools/docs/echo/src/index.ts`, then run `runx tool build --all` to regenerate `manifest.json` and `run.mjs`. Add fixtures in `tools///fixtures/` to lock behaviour. - -Packet IDs are immutable. Schema changes mean a new packet ID, not an in-place edit. - -## Bootstrap - -- Canonical: `runx new docs-demo` -- Cold start: `npm create @runxhq/skill@latest docs-demo` +Edit `run.mjs` to do the real work, and keep both harness classes in `X.yaml`: +one happy path and one stop, error, or refusal case. ## Publish -The scaffold includes `.github/workflows/publish.yml`, which publishes with npm provenance from GitHub Actions. Before publishing, update `package.json` metadata for your repo and package. +```bash +runx login --provider github --for publish +runx registry publish . # the registry runs the harness as the publish gate +``` diff --git a/fixtures/scaffold/new-docs-demo/files/SKILL.md b/fixtures/scaffold/new-docs-demo/files/SKILL.md index 267b36dee..82202f807 100644 --- a/fixtures/scaffold/new-docs-demo/files/SKILL.md +++ b/fixtures/scaffold/new-docs-demo/files/SKILL.md @@ -1,6 +1,29 @@ --- name: docs-demo -description: Scaffolded runx skill package. +description: docs-demo runx skill. Replace this with what the skill does and returns. +source: + type: cli-tool + command: node + args: + - run.mjs + timeout_seconds: 30 + sandbox: + profile: readonly + cwd_policy: skill-directory +inputs: + message: + type: string + required: true + description: Input the skill acts on. Replace with the real inputs. +runx: + category: ops + input_resolution: + required: + - message --- -Use this skill to demonstrate a governed runx authoring package. +# docs-demo + +Describe what this skill does, when an agent should reach for it, and what it +returns. Replace the echo in `run.mjs` with the real work, and add cases to +`X.yaml` so the behaviour is locked by the harness. diff --git a/fixtures/scaffold/new-docs-demo/files/X.yaml b/fixtures/scaffold/new-docs-demo/files/X.yaml index e3def5349..302a4c0a5 100644 --- a/fixtures/scaffold/new-docs-demo/files/X.yaml +++ b/fixtures/scaffold/new-docs-demo/files/X.yaml @@ -1,18 +1,46 @@ skill: docs-demo +version: "0.1.0" + +catalog: + kind: skill + audience: public + visibility: public + role: canonical + +harness: + cases: + - name: docs-demo-smoke + runner: default + inputs: + message: hello + expect: + status: sealed + receipt: + schema: runx.receipt.v1 + state: sealed + disposition: closed + reason_code: process_closed + - name: docs-demo-empty-message-fails + runner: default + inputs: + message: "" + expect: + status: failure + receipt: + schema: runx.receipt.v1 + state: sealed + disposition: closed + reason_code: process_failed runners: default: default: true - type: graph + type: cli-tool + command: node + args: + - run.mjs inputs: message: type: string - required: false - default: hello - graph: - name: docs-demo - steps: - - id: echo - tool: docs.echo - inputs: - message: inputs.message + required: true + description: Input the skill acts on. diff --git a/fixtures/scaffold/new-docs-demo/files/dist/packets/echo.v1.schema.json b/fixtures/scaffold/new-docs-demo/files/dist/packets/echo.v1.schema.json deleted file mode 100644 index 338e48f7a..000000000 --- a/fixtures/scaffold/new-docs-demo/files/dist/packets/echo.v1.schema.json +++ /dev/null @@ -1,15 +0,0 @@ -{ - "$schema": "https://json-schema.org/draft/2020-12/schema", - "$id": "https://schemas.runx.dev/docs/demo/echo/v1.json", - "x-runx-packet-id": "docs.demo.echo.v1", - "type": "object", - "required": [ - "message" - ], - "properties": { - "message": { - "type": "string" - } - }, - "additionalProperties": false -} diff --git a/fixtures/scaffold/new-docs-demo/files/fixtures/agent.replay.json b/fixtures/scaffold/new-docs-demo/files/fixtures/agent.replay.json deleted file mode 100644 index a6aaec708..000000000 --- a/fixtures/scaffold/new-docs-demo/files/fixtures/agent.replay.json +++ /dev/null @@ -1,22 +0,0 @@ -{ - "schema": "runx.replay.v1", - "fixture": "echo-agent-replay", - "prompt_fingerprint": "sha256:31db40e2189f146c20e995cf583b1bb2b1df46f0a2d06f14af10e39b9afbf1e2", - "recorded_at": "1970-01-01T00:00:00.000Z", - "target": { - "kind": "skill", - "ref": "." - }, - "status": "sealed", - "outputs": { - "echo_packet": { - "schema": "docs.demo.echo.v1", - "data": { - "message": "hello" - } - } - }, - "usage": { - "mode": "scaffold" - } -} diff --git a/fixtures/scaffold/new-docs-demo/files/fixtures/agent.yaml b/fixtures/scaffold/new-docs-demo/files/fixtures/agent.yaml deleted file mode 100644 index ca0f13757..000000000 --- a/fixtures/scaffold/new-docs-demo/files/fixtures/agent.yaml +++ /dev/null @@ -1,14 +0,0 @@ -name: echo-agent-replay -lane: agent -target: - kind: skill - ref: . -inputs: - message: hello -agent: - mode: replay -expect: - status: sealed - outputs: - echo_packet: - matches_packet: docs.demo.echo.v1 diff --git a/fixtures/scaffold/new-docs-demo/files/fixtures/repos/readme-only/README.md b/fixtures/scaffold/new-docs-demo/files/fixtures/repos/readme-only/README.md deleted file mode 100644 index 051582016..000000000 --- a/fixtures/scaffold/new-docs-demo/files/fixtures/repos/readme-only/README.md +++ /dev/null @@ -1 +0,0 @@ -# docs-demo diff --git a/fixtures/scaffold/new-docs-demo/files/package.json b/fixtures/scaffold/new-docs-demo/files/package.json deleted file mode 100644 index e264ac84f..000000000 --- a/fixtures/scaffold/new-docs-demo/files/package.json +++ /dev/null @@ -1,27 +0,0 @@ -{ - "name": "docs-demo", - "version": "0.1.0", - "description": "Scaffolded runx skill package.", - "type": "module", - "publishConfig": { - "access": "public" - }, - "scripts": { - "build": "runx tool build --all --json", - "runx:list": "runx list --json", - "runx:doctor": "runx doctor --json", - "runx:dev": "runx dev --lane deterministic --json", - "prepublishOnly": "runx tool build --all --json && runx doctor --json" - }, - "runx": { - "packets": [ - "./dist/packets/*.schema.json" - ] - }, - "devDependencies": { - "@runxhq/authoring": "^0.1.4", - "@runxhq/cli": "^0.5.22", - "@tsconfig/node20": "^20.1.6", - "tsx": "^4.20.6" - } -} diff --git a/fixtures/scaffold/new-docs-demo/files/run.mjs b/fixtures/scaffold/new-docs-demo/files/run.mjs new file mode 100644 index 000000000..0c4e3016a --- /dev/null +++ b/fixtures/scaffold/new-docs-demo/files/run.mjs @@ -0,0 +1,8 @@ +// Inputs arrive as RUNX_INPUT_ environment variables. Do the work and +// write the result to stdout. Replace this echo with the real logic. +const message = process.env.RUNX_INPUT_MESSAGE ?? ""; +if (message.trim().length === 0) { + process.stderr.write("message is required\n"); + process.exit(64); +} +process.stdout.write(`${message}\n`); diff --git a/fixtures/scaffold/new-docs-demo/files/src/packets/echo.ts b/fixtures/scaffold/new-docs-demo/files/src/packets/echo.ts deleted file mode 100644 index 2d27b6dcc..000000000 --- a/fixtures/scaffold/new-docs-demo/files/src/packets/echo.ts +++ /dev/null @@ -1,8 +0,0 @@ -import { definePacket, t } from "@runxhq/authoring"; - -export const EchoPacket = definePacket({ - id: "docs.demo.echo.v1", - schema: t.Object({ - message: t.String(), - }), -}); diff --git a/fixtures/scaffold/new-docs-demo/files/tools/docs/echo/fixtures/basic.yaml b/fixtures/scaffold/new-docs-demo/files/tools/docs/echo/fixtures/basic.yaml deleted file mode 100644 index 7b477d7b5..000000000 --- a/fixtures/scaffold/new-docs-demo/files/tools/docs/echo/fixtures/basic.yaml +++ /dev/null @@ -1,14 +0,0 @@ -name: echo-basic -lane: deterministic -target: - kind: tool - ref: docs.echo -inputs: - message: hello -expect: - status: sealed - output: - subset: - schema: docs.demo.echo.v1 - data: - message: hello diff --git a/fixtures/scaffold/new-docs-demo/files/tools/docs/echo/manifest.json b/fixtures/scaffold/new-docs-demo/files/tools/docs/echo/manifest.json deleted file mode 100644 index ec1a556b9..000000000 --- a/fixtures/scaffold/new-docs-demo/files/tools/docs/echo/manifest.json +++ /dev/null @@ -1,41 +0,0 @@ -{ - "schema": "runx.tool.manifest.v1", - "name": "docs.echo", - "version": "0.1.0", - "description": "Echo a docs message.", - "source": { - "type": "cli-tool", - "command": "node", - "args": [ - "./run.mjs" - ] - }, - "runtime": { - "command": "node", - "args": [ - "./run.mjs" - ] - }, - "inputs": { - "message": { - "type": "string", - "required": false, - "default": "hello" - } - }, - "output": { - "packet": "docs.demo.echo.v1", - "wrap_as": "echo_packet" - }, - "scopes": [ - "docs.read" - ], - "runx": { - "artifacts": { - "wrap_as": "echo_packet" - } - }, - "source_hash": "sha256:43323caad0616b9c0bf771663ac556a6aea2971d65c4e23a59d440d9b0b61229", - "schema_hash": "sha256:d5c0e413e7484e04bec267def5ecfe1f63fafb94d8cd96c7fab17d2608b0631a", - "toolkit_version": "0.1.4" -} diff --git a/fixtures/scaffold/new-docs-demo/files/tools/docs/echo/run.mjs b/fixtures/scaffold/new-docs-demo/files/tools/docs/echo/run.mjs deleted file mode 100644 index 6a8c8f60d..000000000 --- a/fixtures/scaffold/new-docs-demo/files/tools/docs/echo/run.mjs +++ /dev/null @@ -1,6 +0,0 @@ -const fs = require("node:fs"); -const rawInputs = process.env.RUNX_INPUTS_PATH - ? fs.readFileSync(process.env.RUNX_INPUTS_PATH, "utf8") - : (process.env.RUNX_INPUTS_JSON || "{}"); -const inputs = JSON.parse(rawInputs); -process.stdout.write(JSON.stringify({ schema: "docs.demo.echo.v1", data: { message: String(inputs.message || "hello") } })); diff --git a/fixtures/scaffold/new-docs-demo/files/tools/docs/echo/src/index.ts b/fixtures/scaffold/new-docs-demo/files/tools/docs/echo/src/index.ts deleted file mode 100644 index 1ff7e0f9b..000000000 --- a/fixtures/scaffold/new-docs-demo/files/tools/docs/echo/src/index.ts +++ /dev/null @@ -1,18 +0,0 @@ -import { defineTool, stringInput } from "@runxhq/authoring"; - -export default defineTool({ - name: "docs.echo", - version: "0.1.0", - description: "Echo a docs message.", - inputs: { - message: stringInput({ default: "hello" }), - }, - output: { - packet: "docs.demo.echo.v1", - wrap_as: "echo_packet", - }, - scopes: ["docs.read"], - run({ inputs }) { - return { message: inputs.message }; - }, -}); diff --git a/fixtures/scaffold/new-docs-demo/files/tsconfig.json b/fixtures/scaffold/new-docs-demo/files/tsconfig.json deleted file mode 100644 index 0e8456a3d..000000000 --- a/fixtures/scaffold/new-docs-demo/files/tsconfig.json +++ /dev/null @@ -1,12 +0,0 @@ -{ - "extends": "@tsconfig/node20/tsconfig.json", - "compilerOptions": { - "module": "NodeNext", - "moduleResolution": "NodeNext", - "strict": true - }, - "include": [ - "src/**/*.ts", - "tools/**/*.ts" - ] -} diff --git a/fixtures/scaffold/new-docs-demo/manifest.json b/fixtures/scaffold/new-docs-demo/manifest.json index 09f64c427..1351f9820 100644 --- a/fixtures/scaffold/new-docs-demo/manifest.json +++ b/fixtures/scaffold/new-docs-demo/manifest.json @@ -1,29 +1,15 @@ { "name": "docs-demo", - "packet_namespace": "docs.demo", "files": [ - "package.json", - "README.md", "SKILL.md", "X.yaml", - "src/packets/echo.ts", - "dist/packets/echo.v1.schema.json", - "tools/docs/echo/src/index.ts", - "tools/docs/echo/run.mjs", - "tools/docs/echo/manifest.json", - "tools/docs/echo/fixtures/basic.yaml", - "fixtures/agent.yaml", - "fixtures/agent.replay.json", - "fixtures/repos/readme-only/README.md", - ".github/workflows/publish.yml", - ".gitignore", - ".gitattributes", - "tsconfig.json" + "run.mjs", + "README.md", + ".gitignore" ], "next_steps": [ "cd ", - "pnpm install", - "pnpm build", - "runx dev" + "runx harness . --json", + "runx skill . --input message=hello --json" ] } diff --git a/package.json b/package.json index ef8ac4860..c3eb94b32 100644 --- a/package.json +++ b/package.json @@ -22,6 +22,9 @@ "cli:link-global": "pnpm build && node scripts/link-global-cli.mjs", "cli:unlink-global": "node scripts/link-global-cli.mjs --unlink", "cli:check-global": "node scripts/link-global-cli.mjs --check", + "cli:link-dev-native": "cargo build --manifest-path crates/Cargo.toml -p runx-cli && node scripts/link-dev-native-cli.mjs", + "cli:check-dev-native": "node scripts/link-dev-native-cli.mjs --check", + "cli:unlink-dev-native": "node scripts/link-dev-native-cli.mjs --unlink", "dogfood:native-core": "sh scripts/dogfood-native-core.sh", "dogfood:core-skills": "node scripts/dogfood-core-skills.mjs", "dogfood:github-issue-to-pr": "node scripts/dogfood-github-issue-to-pr.mjs", @@ -30,6 +33,7 @@ "docs:api": "tsx scripts/gen-api-index.ts", "docs:exit-codes": "tsx scripts/check-cli-exit-codes.ts", "authoring:check-package-contract": "node scripts/check-authoring-package-contract.mjs", + "release:version:check": "tsx scripts/set-release-version.ts --check", "release:smoke-live": "node scripts/smoke-released-cli-live.mjs", "typecheck": "tsc -p tsconfig.typecheck.json --noEmit", "test": "node scripts/test-workspace.mjs", diff --git a/packages/cli/package.json b/packages/cli/package.json index 6ef691b97..74a622125 100644 --- a/packages/cli/package.json +++ b/packages/cli/package.json @@ -1,6 +1,6 @@ { "name": "@runxhq/cli", - "version": "0.6.0", + "version": "0.6.6", "description": "Runx CLI - native governed runtime for agent skills, tools, graphs, and packets.", "private": false, "license": "MIT", diff --git a/packages/cli/src/args.ts b/packages/cli/src/args.ts index 5b1b138cb..b4607a748 100644 --- a/packages/cli/src/args.ts +++ b/packages/cli/src/args.ts @@ -7,6 +7,7 @@ export interface ParsedArgs { readonly subcommand?: string; readonly mcpAction?: "serve"; readonly mcpRefs?: readonly string[]; + readonly mcpNativeArgs?: readonly string[]; readonly doctorPath?: string; readonly doctorFix: boolean; readonly doctorExplainId?: string; @@ -249,6 +250,7 @@ export function parseArgs(argv: readonly string[]): ParsedArgs { subcommand: positionals[0], mcpAction: isMcp && positionals[0] === "serve" ? "serve" : undefined, mcpRefs: isMcp && positionals[0] === "serve" ? positionals.slice(1) : undefined, + mcpNativeArgs: isMcp && positionals[0] === "serve" ? [command, ...rest] : undefined, doctorPath: isDoctor ? positionals[0] : undefined, doctorFix: isDoctor && truthyFlag(inputs.fix), doctorExplainId: isDoctor && typeof inputs.explain === "string" && inputs.explain !== "true" ? inputs.explain : undefined, diff --git a/packages/cli/src/cli-parser/index.test.ts b/packages/cli/src/cli-parser/index.test.ts new file mode 100644 index 000000000..9c2648749 --- /dev/null +++ b/packages/cli/src/cli-parser/index.test.ts @@ -0,0 +1,101 @@ +import { describe, expect, it } from "vitest"; + +import { + parseRunnerManifestYaml, + parseToolManifestYaml, + SkillParseError, + SkillValidationError, + validateRunnerManifest, +} from "./index.js"; + +describe("CLI runner manifest parser", () => { + it("rejects execution profile YAML references and document markers", () => { + for (const yaml of [ + "---\nskill: example", + "runners:\n one:\n outputs: &shared\n result: string\n", + "runners:\n one:\n outputs: *shared\n", + "runners:\n one:\n runx:\n <<: *shared\n", + "runners:\n one:\n type: !custom graph\n", + ]) { + expect(() => parseRunnerManifestYaml(yaml), yaml).toThrow(SkillParseError); + } + }); + + it("rejects duplicate execution profile mapping keys", () => { + expect(() => parseRunnerManifestYaml(` +runners: + one: + type: agent + type: graph +`)).toThrow(/duplicate mapping key/); + }); + + it("rejects unknown top-level and runner fields", () => { + expect(() => validateRunnerManifest(parseRunnerManifestYaml(` +skill: example +unexpected: true +runners: + one: + type: agent +`))).toThrow(SkillValidationError); + + expect(() => validateRunnerManifest(parseRunnerManifestYaml(` +runners: + one: + type: agent + typo_field: true +`))).toThrow(SkillValidationError); + }); + + it("accepts the governed act effect source fields", () => { + const manifest = validateRunnerManifest(parseRunnerManifestYaml(` +version: "1" +runners: + observe: + type: http + url: https://example.test/observe + method: POST + act: + effect_field_from: effect_field + effect_from_input: thread_locator + effect_prefix_from: effect_prefix +`)); + + expect(manifest.version).toBe("1"); + expect(manifest.runners.observe?.source.act).toMatchObject({ + effect_field_from: "effect_field", + effect_from_input: "thread_locator", + effect_prefix_from: "effect_prefix", + }); + }); + + it("normalizes nested http source declarations", () => { + const manifest = validateRunnerManifest(parseRunnerManifestYaml(` +runners: + fetch: + source: + type: http + http: + url: https://example.test/api + method: GET + headers: + accept: application/json + allow_private_network: false +`)); + + expect(manifest.runners.fetch?.source.http).toEqual({ + url: "https://example.test/api", + method: "GET", + headers: { + accept: "application/json", + }, + allowPrivateNetwork: false, + }); + }); +}); + +describe("CLI tool manifest parser", () => { + it("keeps YAML parity checks for tool manifests", () => { + expect(() => parseToolManifestYaml("name: tool\ndescription: one: two\n")).toThrow(SkillParseError); + }); +}); diff --git a/packages/cli/src/cli-parser/index.ts b/packages/cli/src/cli-parser/index.ts index e6e1d26f5..6bb802dcd 100644 --- a/packages/cli/src/cli-parser/index.ts +++ b/packages/cli/src/cli-parser/index.ts @@ -1,6 +1,11 @@ import { parseDocument } from "yaml"; import { validateGraphDocument, type ExecutionGraph } from "./graph.js"; +import { + assertExecutionProfileYamlSubset, + assertYamlParitySubset, + YamlSubsetError, +} from "./yaml-subset.js"; import { normalizeSandboxDeclaration } from "../cli-sandbox.js"; import { GOVERNED_DISPOSITIONS, type ExecutionSemantics } from "../cli-execution-semantics.js"; import { errorMessage, isRecord, readField } from "../cli-util.js"; @@ -53,9 +58,41 @@ export interface SkillSource { readonly hook?: string; readonly outputs?: Readonly>; readonly graph?: ExecutionGraph; + readonly http?: SkillHttpSource; + readonly act?: ActDeclaration; readonly raw: Record; } +export interface SkillHttpSource { + readonly url: string; + readonly method?: string; + readonly headers?: Readonly>; + readonly allowPrivateNetwork?: boolean; +} + +export interface ActDeclaration { + readonly form?: string; + readonly form_from?: string; + readonly purpose?: string; + readonly purpose_from?: string; + readonly legitimacy?: string; + readonly legitimacy_from?: string; + readonly reason_from?: string; + readonly target_from?: string; + readonly decision_from?: string; + readonly effect_from?: string; + readonly effect_field_from?: string; + readonly effect_from_input?: string; + readonly effect_type?: string; + readonly effect_prefix?: string; + readonly effect_prefix_from?: string; + readonly actor_from?: string; + readonly authority_from?: string; + readonly previous_from?: string; + readonly reason_step?: string; + readonly effect_step?: string; +} + export interface SkillArtifactContract { readonly emits?: readonly string[]; readonly namedEmits?: Readonly>; @@ -83,6 +120,8 @@ export interface SkillSandbox { export interface ValidatedSkill { readonly name: string; readonly description?: string; + readonly category?: string; + readonly runxCategory?: string; readonly body: string; readonly source: SkillSource; readonly inputs: Readonly>; @@ -131,7 +170,7 @@ export interface SkillRunnerDefinition { export type PostRunReflectPolicy = "auto" | "always" | "never"; export type CatalogKind = "skill" | "graph"; -export type CatalogAudience = "public" | "builder" | "operator"; +export type CatalogAudience = "public" | "builder" | "operator" | "system"; export type CatalogVisibility = "public" | "internal"; export type CatalogRole = | "canonical" @@ -194,6 +233,10 @@ export interface RunnerHarnessManifest { export interface SkillRunnerManifest { readonly skill?: string; + readonly version?: string; + readonly runx?: Readonly>; + readonly policy?: unknown; + readonly emits?: unknown; readonly catalog?: CatalogMetadata; readonly runners: Readonly>; readonly harness?: RunnerHarnessManifest; @@ -259,6 +302,7 @@ export function parseSkillMarkdown(markdown: string): RawSkillIR { } export function parseRunnerManifestYaml(yaml: string): RawRunnerManifestIR { + assertYamlSubset("runner_manifest", yaml, "execution-profile"); const document = parseDocument(yaml, { prettyErrors: false }); if (document.errors.length > 0) { throw new SkillParseError(document.errors.map((error) => error.message).join("; ")); @@ -276,6 +320,7 @@ export function parseRunnerManifestYaml(yaml: string): RawRunnerManifestIR { } export function parseToolManifestYaml(yaml: string): RawToolManifestIR { + assertYamlSubset("tool_manifest", yaml, "parity"); const document = parseDocument(yaml, { prettyErrors: false }); if (document.errors.length > 0) { throw new SkillParseError(document.errors.map((error) => error.message).join("; ")); @@ -326,11 +371,15 @@ export function validateSkill(raw: RawSkillIR, options: ValidateSkillOptions = { } const source = validateSource(sourceRecord ?? { type: "agent" }, isRecord(runxValue) ? runxValue : undefined); const runx = isRecord(runxValue) ? runxValue : undefined; + const category = validatePortableSkillCategory(raw.frontmatter.category); + const runxCategory = validateRunxSkillCategory(readField(runx, "category")); const risk = raw.frontmatter.risk; return { name, description, + category, + runxCategory, body: raw.body, source, inputs, @@ -352,6 +401,19 @@ export function validateSkill(raw: RawSkillIR, options: ValidateSkillOptions = { }; } +function validatePortableSkillCategory(value: unknown): string | undefined { + return normalizeOptionalCategory(optionalNullableString(value, "category")); +} + +function validateRunxSkillCategory(value: unknown): string | undefined { + return normalizeOptionalCategory(optionalNullableString(value, "runx.category")); +} + +function normalizeOptionalCategory(value: string | undefined): string | undefined { + const normalized = value?.trim(); + return normalized ? normalized : undefined; +} + export function extractSkillQualityProfile(body: string): SkillQualityProfile | undefined { const content = extractMarkdownSection(body, "Quality Profile", 2); if (!content) { @@ -365,13 +427,18 @@ export function extractSkillQualityProfile(body: string): SkillQualityProfile | export function validateRunnerManifest(raw: RawRunnerManifestIR): SkillRunnerManifest { const runnersRecord = requiredNullableRecord(raw.document.runners, "runners"); + rejectUnknownFields(raw.document, "runner_manifest", ["skill", "version", "runx", "policy", "emits", "catalog", "runners", "harness"]); const runners: Record = {}; for (const [name, value] of Object.entries(runnersRecord)) { const runner = requiredNullableRecord(value, `runners.${name}`); + rejectUnknownFields(runner, `runners.${name}`, runnerFields); const runx = optionalNullableRecord(runner.runx, `runners.${name}.runx`); validatePostRunReflectPolicy(runx, `runners.${name}.runx`); const sourceRecord = optionalNullableRecord(runner.source, `runners.${name}.source`) ?? runner; + if (runner.source !== undefined) { + rejectUnknownFields(sourceRecord, `runners.${name}.source`, sourceFields); + } const risk = runner.risk; runners[name] = { name, @@ -407,6 +474,10 @@ export function validateRunnerManifest(raw: RawRunnerManifestIR): SkillRunnerMan return { skill: optionalNullableString(raw.document.skill, "skill"), + version: optionalNullableString(raw.document.version, "version"), + runx: optionalNullableRecord(raw.document.runx, "runx"), + policy: raw.document.policy, + emits: raw.document.emits, catalog: validateCatalogMetadata(optionalNullableRecord(raw.document.catalog, "catalog"), "catalog"), runners, harness, @@ -430,8 +501,8 @@ function validateCatalogMetadata(value: Record | undefined, lab if (kind !== "skill" && kind !== "graph") { throw new SkillValidationError(`${label}.kind must be skill or graph.`); } - if (audience !== "public" && audience !== "builder" && audience !== "operator") { - throw new SkillValidationError(`${label}.audience must be public, builder, or operator.`); + if (audience !== "public" && audience !== "builder" && audience !== "operator" && audience !== "system") { + throw new SkillValidationError(`${label}.audience must be public, builder, operator, or system.`); } if (visibility !== "public" && visibility !== "internal") { throw new SkillValidationError(`${label}.visibility must be public or internal.`); @@ -567,6 +638,7 @@ export function resolvePostRunReflectPolicy( function validateSource(source: Record, runx: Record | undefined): SkillSource { const type = requiredNullableString(source.type, "source.type"); + validateSourceType(type, "source.type"); const args = optionalNullableStringArray(source.args, "source.args") ?? []; const inputMode = optionalInputMode(source.input_mode); const timeoutSeconds = optionalNullableNumber(source.timeout_seconds, "source.timeout_seconds"); @@ -596,6 +668,8 @@ function validateSource(source: Record, runx: Record, runx: Record, type: string): SkillHttpSource | undefined { + if (type !== "http") { + return undefined; + } + const http = optionalNullableRecord(source.http, "source.http") ?? source; + return { + url: requiredNullableString(http.url, "source.url"), + method: validateHttpMethod(optionalNullableString(http.method, "source.method")), + headers: validateHttpHeaders(http.headers), + allowPrivateNetwork: optionalNullableBoolean(http.allow_private_network, "source.allow_private_network"), + }; +} + +function validateHttpMethod(method: string | undefined): string | undefined { + if (method === undefined) { + return undefined; + } + if (["GET", "POST", "PUT", "PATCH", "DELETE"].includes(method.toUpperCase())) { + return method; + } + throw new SkillValidationError(`source.method ${method} is not supported; use GET, POST, PUT, PATCH, or DELETE.`); +} + +function validateHttpHeaders(value: unknown): Readonly> | undefined { + const headers = optionalNullableRecord(value, "source.headers"); + if (!headers) { + return undefined; + } + for (const [key, entry] of Object.entries(headers)) { + if (typeof entry !== "string") { + throw new SkillValidationError(`source.headers.${key} must be a string.`); + } + } + return headers as Readonly>; +} + +const actFields = [ + "form", + "form_from", + "purpose", + "purpose_from", + "legitimacy", + "legitimacy_from", + "reason_from", + "target_from", + "decision_from", + "effect_from", + "effect_field_from", + "effect_from_input", + "effect_type", + "effect_prefix", + "effect_prefix_from", + "actor_from", + "authority_from", + "previous_from", + "reason_step", + "effect_step", +] as const; + +function validateActDeclaration(value: unknown, field: string): ActDeclaration | undefined { + const record = optionalNullableRecord(value, field); + if (!record) { + return undefined; + } + rejectUnknownFields(record, field, actFields); + const validated: Record = {}; + for (const key of actFields) { + const entry = optionalNullableString(record[key], `${field}.${key}`); + if (entry !== undefined) { + validated[key] = entry; + } + } + return validated; +} + function validateSandbox(value: unknown): SkillSandbox | undefined { if (value === undefined || value === null) { return undefined; @@ -1095,5 +1253,101 @@ function optionalCwdPolicy(value: unknown): SkillSandbox["cwdPolicy"] { throw new SkillValidationError("sandbox.cwd_policy must be skill-directory, workspace, or custom."); } +function rejectUnknownFields( + record: Record, + field: string, + allowed: readonly string[], +): void { + for (const key of Object.keys(record)) { + if (!allowed.includes(key)) { + throw new SkillValidationError(`${field}.${key} is not supported; allowed fields: ${allowed.join(", ")}.`); + } + } +} + +function assertYamlSubset(field: string, yaml: string, kind: "execution-profile" | "parity"): void { + try { + if (kind === "execution-profile") { + assertExecutionProfileYamlSubset(field, yaml); + } else { + assertYamlParitySubset(field, yaml); + } + } catch (error) { + if (error instanceof YamlSubsetError) { + throw new SkillParseError(error.message, { cause: error }); + } + throw error; + } +} + +const sourceTypes = [ + "cli-tool", + "mcp", + "catalog", + "a2a", + "agent", + "agent-task", + "harness-hook", + "graph", + "http", + "external-adapter", + "thread-outbox-provider", +] as const; + +const sourceFields = [ + "act", + "agent", + "agent_card_url", + "agent_identity", + "allow_private_network", + "args", + "arguments", + "catalog_ref", + "command", + "cwd", + "external_adapter", + "external_adapter_manifest", + "external_adapter_manifest_path", + "graph", + "headers", + "hook", + "http", + "input_mode", + "invocation_id", + "method", + "outputs", + "run_id", + "sandbox", + "server", + "skill_ref", + "task", + "timeout_seconds", + "tool", + "type", + "url", +] as const; + +const runnerFields = [ + ...sourceFields, + "allowed_tools", + "artifacts", + "auth", + "context", + "context_skills", + "default", + "execution", + "idempotency", + "inputs", + "instructions", + "mutating", + "policy", + "retry", + "risk", + "runx", + "runtime", + "scopes", + "source", +] as const; + export * from "./graph.js"; diff --git a/packages/cli/src/cli-parser/yaml-subset.ts b/packages/cli/src/cli-parser/yaml-subset.ts new file mode 100644 index 000000000..e13b11f46 --- /dev/null +++ b/packages/cli/src/cli-parser/yaml-subset.ts @@ -0,0 +1,393 @@ +const divergentBoolish = ["yes", "no", "on", "off"] as const; + +export class YamlSubsetError extends Error { + constructor(message: string) { + super(message); + this.name = "YamlSubsetError"; + } +} + +export function assertYamlParitySubset(field: string, source: string): void { + let blockScalarIndent: number | undefined; + for (const [lineIndex, line] of source.split(/\r?\n/).entries()) { + const lineNumber = lineIndex + 1; + const content = stripYamlComment(line); + if (content === undefined) { + continue; + } + const trimmed = content.trim(); + if (blockScalarIndent !== undefined) { + if (trimmed === "" || leadingSpaces(content) > blockScalarIndent) { + continue; + } + blockScalarIndent = undefined; + } + if (trimmed === "" || trimmed.startsWith("---") || trimmed.startsWith("...")) { + continue; + } + rejectExplicitMappingKey(field, lineNumber, trimmed); + rejectEmbeddedColonKey(field, lineNumber, trimmed); + rejectColonSpacePlainScalar(field, lineNumber, content); + blockScalarIndent = blockScalarIndentAfter(content) ?? blockScalarIndent; + } +} + +export function assertExecutionProfileYamlSubset(field: string, source: string): void { + assertYamlParitySubset(field, source); + const mappingStack: MappingFrame[] = []; + let blockScalarIndent: number | undefined; + for (const [lineIndex, line] of source.split(/\r?\n/).entries()) { + const lineNumber = lineIndex + 1; + const content = stripYamlComment(line); + if (content === undefined) { + continue; + } + const trimmed = content.trim(); + if (blockScalarIndent !== undefined) { + if (trimmed === "" || leadingSpaces(content) > blockScalarIndent) { + continue; + } + blockScalarIndent = undefined; + } + if (trimmed === "") { + continue; + } + rejectDocumentMarker(field, lineNumber, trimmed); + rejectYamlReferenceSyntax(field, lineNumber, content); + rejectDuplicateMappingKey(field, lineNumber, content, mappingStack); + blockScalarIndent = blockScalarIndentAfter(content) ?? blockScalarIndent; + } +} + +export function yamlScalarSubsetAllows(literal: string): boolean { + const trimmed = literal.trim(); + return !isBoolish(trimmed) + && !isBasePrefixedNumber(trimmed) + && !isSexagesimalLike(trimmed) + && !isDateLike(trimmed) + && !isSpecialFloat(trimmed); +} + +export function assertYamlScalarSubset(field: string, literal: string): void { + if (yamlScalarSubsetAllows(literal)) { + return; + } + throw new YamlSubsetError(`${field} uses unsupported YAML scalar ${JSON.stringify(literal)}.`); +} + +function stripYamlComment(line: string): string | undefined { + const scanner = new QuoteScanner(); + for (let index = 0; index < line.length; index += 1) { + const char = line[index]!; + if (scanner.isPlainAt(char) && char === "#" && isCommentStart(line, index)) { + return line.slice(0, index); + } + scanner.consume(char); + } + return line; +} + +function isCommentStart(line: string, index: number): boolean { + return index === 0 || /\s/.test(line[index - 1]!); +} + +function rejectExplicitMappingKey(field: string, lineNumber: number, trimmed: string): void { + if (trimmed === "?" || trimmed.startsWith("? ")) { + throw ambiguousYaml(field, lineNumber, trimmed); + } +} + +function rejectEmbeddedColonKey(field: string, lineNumber: number, trimmed: string): void { + const key = topLevelPlainKey(trimmed)?.[0]; + if (key?.includes(":")) { + throw ambiguousYaml(field, lineNumber, trimmed); + } +} + +function topLevelPlainKey(trimmed: string): [string, number] | undefined { + const first = trimmed[0]; + if (first === undefined || ["-", "?", "{", "[", "\"", "'"].includes(first)) { + return undefined; + } + const scanner = new QuoteScanner(); + for (let index = 0; index < trimmed.length; index += 1) { + const char = trimmed[index]!; + if (scanner.isPlainAt(char) && char === ":" && isMappingDelimiter(trimmed, index)) { + return [trimmed.slice(0, index).trim(), index]; + } + scanner.consume(char); + } + return undefined; +} + +type QuoteState = + | "plain" + | "in-double" + | "in-single-pending-apostrophe" + | "in-single" + | "in-double-escape"; + +class QuoteScanner { + private state: QuoteState = "plain"; + + isPlainAt(char: string): boolean { + if (this.state === "plain") { + return true; + } + if (this.state === "in-single-pending-apostrophe") { + return char !== "'"; + } + return false; + } + + consume(char: string): void { + if (this.state === "plain") { + this.state = this.plainStateAfter(char); + return; + } + if (this.state === "in-double") { + this.state = char === "\\" ? "in-double-escape" : char === "\"" ? "plain" : "in-double"; + return; + } + if (this.state === "in-double-escape") { + this.state = "in-double"; + return; + } + if (this.state === "in-single") { + this.state = char === "'" ? "in-single-pending-apostrophe" : "in-single"; + return; + } + this.state = char === "'" ? "in-single" : this.plainStateAfter(char); + } + + private plainStateAfter(char: string): QuoteState { + if (char === "'") { + return "in-single"; + } + if (char === "\"") { + return "in-double"; + } + return "plain"; + } +} + +function isMappingDelimiter(value: string, index: number): boolean { + const next = value[index + 1]; + return next === undefined || /\s/.test(next); +} + +function rejectColonSpacePlainScalar(field: string, lineNumber: number, content: string): void { + const split = splitPlainMappingValue(content); + if (!split) { + return; + } + const [, value] = split; + if (plainScalarContainsColonSpace(value)) { + throw ambiguousYaml(field, lineNumber, value.trim()); + } +} + +function rejectDocumentMarker(field: string, lineNumber: number, trimmed: string): void { + if ( + trimmed === "---" + || trimmed === "..." + || trimmed.startsWith("--- ") + || trimmed.startsWith("... ") + ) { + throw new YamlSubsetError( + `${field}: YAML document markers are not supported in X.yaml at line ${lineNumber}; use one plain profile document.`, + ); + } +} + +function rejectYamlReferenceSyntax(field: string, lineNumber: number, content: string): void { + for (const token of [": &", ": *", ": !", "- &", "- *", "- !"] as const) { + if (containsPlainToken(content, token)) { + throw new YamlSubsetError( + `${field}: YAML anchors, aliases, and tags are not supported in X.yaml at line ${lineNumber}; write the profile explicitly.`, + ); + } + } + const trimmed = content.trimStart(); + if (trimmed.startsWith("&") || trimmed.startsWith("*") || trimmed.startsWith("!")) { + throw new YamlSubsetError( + `${field}: YAML anchors, aliases, and tags are not supported in X.yaml at line ${lineNumber}; write the profile explicitly.`, + ); + } +} + +function containsPlainToken(content: string, token: string): boolean { + const scanner = new QuoteScanner(); + for (let index = 0; index < content.length; index += 1) { + const char = content[index]!; + if (scanner.isPlainAt(char) && content.startsWith(token, index)) { + return true; + } + scanner.consume(char); + } + return false; +} + +interface MappingFrame { + readonly indent: number; + readonly keys: Set; +} + +function rejectDuplicateMappingKey( + field: string, + lineNumber: number, + content: string, + stack: MappingFrame[], +): void { + const indent = leadingSpaces(content); + const trimmed = content.trimStart(); + const sequenceKey = sequenceItemKey(trimmed, indent); + const keyMatch = sequenceKey + ? { keyIndent: sequenceKey[0], key: sequenceKey[1], sequenceItem: true } + : topLevelPlainKey(trimmed) + ? { keyIndent: indent, key: topLevelPlainKey(trimmed)![0], sequenceItem: false } + : undefined; + if (!keyMatch) { + return; + } + const { key, keyIndent, sequenceItem } = keyMatch; + if (key === "<<") { + throw new YamlSubsetError( + `${field}: YAML merge keys are not supported in X.yaml at line ${lineNumber}; write the profile explicitly.`, + ); + } + if (sequenceItem) { + while (stack.at(-1) && stack.at(-1)!.indent >= keyIndent) { + stack.pop(); + } + } else { + while (stack.at(-1) && stack.at(-1)!.indent > keyIndent) { + stack.pop(); + } + } + if (!stack.at(-1) || stack.at(-1)!.indent !== keyIndent) { + stack.push({ indent: keyIndent, keys: new Set() }); + } + const frame = stack.at(-1)!; + if (frame.keys.has(key)) { + throw new YamlSubsetError( + `${field}: duplicate mapping key ${JSON.stringify(key)} in X.yaml at line ${lineNumber}; keep profile keys unique.`, + ); + } + frame.keys.add(key); +} + +function blockScalarIndentAfter(content: string): number | undefined { + return blockScalarValueCandidates(content).some(isBlockScalarHeader) ? leadingSpaces(content) : undefined; +} + +function blockScalarValueCandidates(content: string): string[] { + const candidates: string[] = []; + const mapping = splitPlainMappingValue(content); + if (mapping) { + candidates.push(mapping[1]); + } + const trimmed = content.trimStart(); + if (trimmed.startsWith("- ")) { + const item = trimmed.slice(2).trimStart(); + candidates.push(item); + const itemMapping = splitPlainMappingValue(item); + if (itemMapping) { + candidates.push(itemMapping[1]); + } + } + return candidates; +} + +function isBlockScalarHeader(value: string): boolean { + return /^[|>](?:[+-]?\d?|\d?[+-]?)$/.test(value.trim()); +} + +function sequenceItemKey(trimmed: string, indent: number): [number, string] | undefined { + const rest = trimmed.startsWith("- ") ? trimmed.slice(2) : undefined; + if (rest === undefined) { + return undefined; + } + const item = rest.trimStart(); + const leading = rest.length - item.length; + const key = topLevelPlainKey(item)?.[0]; + return key === undefined ? undefined : [indent + 2 + leading, key]; +} + +function leadingSpaces(content: string): number { + return content.length - content.trimStart().length; +} + +function splitPlainMappingValue(content: string): [string, string] | undefined { + const trimmed = content.trimStart(); + const split = topLevelPlainKey(trimmed); + if (!split) { + return undefined; + } + const [key, delimiterIndex] = split; + return [key, trimmed.slice(delimiterIndex + 1)]; +} + +function plainScalarContainsColonSpace(value: string): boolean { + const trimmed = value.trimStart(); + if ( + trimmed === "" + || trimmed.startsWith("\"") + || trimmed.startsWith("'") + || trimmed.startsWith("|") + || trimmed.startsWith(">") + || trimmed.startsWith("{") + || trimmed.startsWith("[") + || trimmed === "null" + || trimmed === "true" + || trimmed === "false" + ) { + return false; + } + return containsUnquotedColonSpace(trimmed); +} + +function containsUnquotedColonSpace(value: string): boolean { + const scanner = new QuoteScanner(); + for (let index = 0; index < value.length; index += 1) { + const char = value[index]!; + if (scanner.isPlainAt(char) && char === ":" && isMappingDelimiter(value, index)) { + return true; + } + scanner.consume(char); + } + return false; +} + +function ambiguousYaml(field: string, lineNumber: number, literal: string): YamlSubsetError { + return new YamlSubsetError( + `${field}: ambiguous YAML construct at line ${lineNumber}; quote the value or key: ${literal}`, + ); +} + +function isBoolish(value: string): boolean { + return divergentBoolish.some((candidate) => candidate.toLowerCase() === value.toLowerCase()); +} + +function isBasePrefixedNumber(value: string): boolean { + const unsigned = value.replace(/^[+-]/, ""); + return unsigned.startsWith("0x") || unsigned.startsWith("0X") || unsigned.startsWith("0o"); +} + +function isSexagesimalLike(value: string): boolean { + const unsigned = value.replace(/^[+-]/, ""); + const parts = unsigned.split(":"); + const [first, ...rest] = parts; + return Boolean(first) + && /^\d+$/.test(first) + && rest.length > 0 + && rest.every((part) => part !== "" && /^\d+$/.test(part)); +} + +function isDateLike(value: string): boolean { + return /^\d{4}-\d{2}-\d{2}/.test(value); +} + +function isSpecialFloat(value: string): boolean { + return [".nan", ".inf", "+.inf", "-.inf"].includes(value.toLowerCase()); +} diff --git a/packages/cli/src/cli-registry.ts b/packages/cli/src/cli-registry.ts index 26532edd1..4c636617e 100644 --- a/packages/cli/src/cli-registry.ts +++ b/packages/cli/src/cli-registry.ts @@ -6,6 +6,8 @@ export interface SkillSearchResult { readonly skill_id: string; readonly name: string; readonly summary?: string; + readonly category?: string; + readonly source_category?: string; readonly owner: string; readonly version?: string; readonly digest?: string; diff --git a/packages/cli/src/cli-runtime-contracts.ts b/packages/cli/src/cli-runtime-contracts.ts index b54b13328..1c3153913 100644 --- a/packages/cli/src/cli-runtime-contracts.ts +++ b/packages/cli/src/cli-runtime-contracts.ts @@ -37,6 +37,8 @@ export interface RegistrySkillVersion { readonly runner_names: readonly string[]; readonly skill_id: string; readonly name: string; + readonly category?: string; + readonly source_category?: string; readonly version: string; readonly digest: string; readonly source_type: string; diff --git a/packages/cli/src/commands/mcp.test.ts b/packages/cli/src/commands/mcp.test.ts index 833fe6a6a..52faef898 100644 --- a/packages/cli/src/commands/mcp.test.ts +++ b/packages/cli/src/commands/mcp.test.ts @@ -5,6 +5,7 @@ import { PassThrough } from "node:stream"; import { describe, expect, it } from "vitest"; +import { parseArgs } from "../args.js"; import { handleMcpServeCommand } from "./mcp.js"; import { resolveRunxBinary } from "../../../../tests/runx-binary.js"; @@ -12,6 +13,30 @@ const workspaceRoot = process.cwd(); const runxBinary = resolveRunxBinary(); describe("runx mcp serve", () => { + it("preserves native argv for Rust-owned MCP flags", () => { + const parsed = parseArgs([ + "mcp", + "serve", + "runx/weather", + "--receipt-dir", + "receipts", + "--http-listen", + "127.0.0.1:3333", + "--http-allow-non-loopback", + ]); + + expect(parsed.mcpNativeArgs).toEqual([ + "mcp", + "serve", + "runx/weather", + "--receipt-dir", + "receipts", + "--http-listen", + "127.0.0.1:3333", + "--http-allow-non-loopback", + ]); + }); + it("lists served skills and executes through the local kernel", async () => { const tempDir = await mkdtemp(path.join(os.tmpdir(), "runx-mcp-serve-")); const skillDir = path.join(tempDir, "echo"); diff --git a/packages/cli/src/commands/mcp.ts b/packages/cli/src/commands/mcp.ts index 6d51a0b0f..4f5ce68bb 100644 --- a/packages/cli/src/commands/mcp.ts +++ b/packages/cli/src/commands/mcp.ts @@ -7,6 +7,7 @@ import type { CliIo } from "../index.js"; export interface McpCommandArgs { readonly mcpRefs?: readonly string[]; + readonly mcpNativeArgs?: readonly string[]; readonly runner?: string; readonly receiptDir?: string; } @@ -37,7 +38,7 @@ export async function handleMcpServeCommand( await runNativeMcpProcess({ command: resolveNativeRunxCommand(env), - args: nativeMcpServeArgs(parsed, skillRefs), + args: parsed.mcpNativeArgs ?? nativeMcpServeArgs(parsed, skillRefs), cwd: env.RUNX_CWD || process.cwd(), env: { ...process.env, @@ -132,4 +133,3 @@ function nativeMcpExitMessage(status: number | null, stderr: string): string { const details = stderr.trim(); return `Native MCP serve failed with exit ${status ?? "unknown"}${details ? `: ${details}` : "."}`; } - diff --git a/packages/cli/src/commands/new.ts b/packages/cli/src/commands/new.ts index 0f746b515..2c4c642ff 100644 --- a/packages/cli/src/commands/new.ts +++ b/packages/cli/src/commands/new.ts @@ -8,9 +8,8 @@ export interface NewCommandArgs { } export interface NewResult { - readonly action: "package"; + readonly action: "skill"; readonly name: string; - readonly packet_namespace: string; readonly directory: string; readonly files: readonly string[]; readonly next_steps: readonly string[]; @@ -26,7 +25,7 @@ export async function handleNewCommand(parsed: NewCommandArgs, env: NodeJS.Proce directory, }); return { - action: "package", + action: "skill", ...result, }; } diff --git a/packages/cli/src/commands/url-add.test.ts b/packages/cli/src/commands/url-add.test.ts index fd8569e1a..67e7cbb79 100644 --- a/packages/cli/src/commands/url-add.test.ts +++ b/packages/cli/src/commands/url-add.test.ts @@ -33,8 +33,8 @@ describe("isGithubRepoUrl", () => { }); describe("resolveUrlAddApiBaseUrl", () => { - it("falls back to runx.ai", () => { - expect(resolveUrlAddApiBaseUrl({})).toBe("https://runx.ai"); + it("falls back to the hosted API origin", () => { + expect(resolveUrlAddApiBaseUrl({})).toBe("https://api.runx.ai"); }); it("respects RUNX_PUBLIC_API_BASE_URL", () => { diff --git a/packages/cli/src/commands/url-add.ts b/packages/cli/src/commands/url-add.ts index 3201924c9..3fcb46b03 100644 --- a/packages/cli/src/commands/url-add.ts +++ b/packages/cli/src/commands/url-add.ts @@ -141,5 +141,5 @@ export function renderUrlAddResult(result: UrlAddIndexResult): string { } export function resolveUrlAddApiBaseUrl(env: Record): string { - return env.RUNX_PUBLIC_API_BASE_URL?.trim() || "https://runx.ai"; + return env.RUNX_PUBLIC_API_BASE_URL?.trim() || "https://api.runx.ai"; } diff --git a/packages/cli/src/help.ts b/packages/cli/src/help.ts index 32820532c..d5b2885bc 100644 --- a/packages/cli/src/help.ts +++ b/packages/cli/src/help.ts @@ -40,26 +40,11 @@ export function writeUsage(stream: Writable, env: NodeJS.ProcessEnv = process.en " runx --help", " runx --version", "", - "Commands:", - " runx new [--directory dir] [--json]", - " runx init [-g|--global] [--prefetch official] [--json]", - " runx history [query] [--skill s] [--status s] [--source s] [--actor a] [--artifact-type t] [--since iso] [--until iso] [--receipt-dir dir] [--json]", - " runx list [tools|skills|graphs|packets|overlays] [--ok-only|--invalid-only] [--json]", - " runx config set|get|list [agent.provider|agent.model|agent.api_key] [value] [--json]", - " runx policy inspect|lint [--json]", - " runx publish [--api-base-url url] [--token token] [--allow-local-api] [--json]", - " runx kernel eval --input --json", - " runx doctor [path] [--json]", - " runx dev [root] [--lane lane] [--json]", - " runx mcp serve [--receipt-dir dir]", - " runx add [--registry url|path] [--version version] [--to dir] [--digest sha256] [--json]", - " runx add [--ref git-ref] [--api-base-url url] [--json]", - " runx skill [--registry url|path] [--digest sha256] [--runner name] [--input key=value] [--flag value] [--receipt-dir dir] [--run-id id --answers file] [--json]", - " runx harness [--json]", - " runx tool build |--all [--json]", - " runx tool search [--source source] [--json]", - " runx tool inspect [--source source] [--json]", - " runx registry search|read|resolve|install|publish ... --json", + "Native help is authoritative:", + " runx --help", + " runx --help", + "", + "This TypeScript entrypoint is a package launcher and test harness only; command grammar lives in the Rust binary.", "", ].join("\n"), ); diff --git a/packages/cli/src/index.test.ts b/packages/cli/src/index.test.ts index 2c6c730af..7aac21a6b 100644 --- a/packages/cli/src/index.test.ts +++ b/packages/cli/src/index.test.ts @@ -127,6 +127,11 @@ Return the provided task id. answers: { "agent_task.child-task.output": { echoed_task: "abc-123", + closure: { + disposition: "closed", + reason_code: "test_answer", + summary: "test answer supplied by caller", + }, }, }, }, @@ -176,7 +181,7 @@ Return the provided task id. expect(secondJson).toMatchObject({ status: "sealed", }); - expect(JSON.parse(secondJson.execution.stdout)).toEqual({ echoed_task: "abc-123" }); + expect(JSON.parse(secondJson.execution.stdout)).toMatchObject({ echoed_task: "abc-123" }); }); it("does not treat arbitrary top-level commands as skill invocations", () => { @@ -365,7 +370,8 @@ Return the provided task id. expect(exitCode).toBe(64); expect(stdout.contents()).toBe(""); expect(stderr.contents()).toContain("Usage:"); - expect(stderr.contents()).toContain("runx skill "); + expect(stderr.contents()).toContain("Native help is authoritative:"); + expect(stderr.contents()).toContain("runx --help"); }); it("routes sourcey through the native graph runner without TS fallback", async () => { @@ -820,6 +826,11 @@ Return the grounded label. answers: { "agent_task.summarize-label.output": { summary: "grounded from caller answer", + closure: { + disposition: "closed", + reason_code: "test_answer", + summary: "test answer supplied by caller", + }, }, }, }, @@ -839,7 +850,7 @@ Return the grounded label. expect(continuedExit).toBe(0); expect(continuedStderr.contents()).toBe(""); const continued = JSON.parse(continuedStdout.contents()) as { execution: { stdout: string } }; - expect(JSON.parse(continued.execution.stdout)).toEqual({ summary: "grounded from caller answer" }); + expect(JSON.parse(continued.execution.stdout)).toMatchObject({ summary: "grounded from caller answer" }); expect(requestCount).toBe(0); }); @@ -1189,7 +1200,7 @@ Answer the prompt directly. expect(globalThis.fetch).not.toHaveBeenCalled(); }); - it("renders top-level help with starter flows and admin commands", async () => { + it("renders top-level help as a native grammar launcher", async () => { const stdout = createMemoryStream(); const stderr = createMemoryStream(); @@ -1197,16 +1208,14 @@ Answer the prompt directly. expect(exitCode).toBe(0); expect(stderr.contents()).toBe(""); - expect(stdout.contents()).toContain("Commands:"); - expect(stdout.contents()).toContain("runx history [query]"); - expect(stdout.contents()).toContain("runx add "); - expect(stdout.contents()).toContain("runx add [--ref git-ref] [--api-base-url url]"); - expect(stdout.contents()).toContain("runx skill "); - expect(stdout.contents()).toContain("runx harness "); - expect(stdout.contents()).toContain("runx tool inspect "); - expect(stdout.contents()).not.toContain("runx evolve"); - expect(stdout.contents()).not.toContain("runx skill inspect "); - expect(stdout.contents()).not.toContain("runx export-receipts --trainable"); + expect(stdout.contents()).toContain("Native help is authoritative:"); + expect(stdout.contents()).toContain("runx --help"); + expect(stdout.contents()).toContain("runx --help"); + expect(stdout.contents()).toContain("command grammar lives in the Rust binary"); + expect(stdout.contents()).not.toContain("Commands:"); + expect(stdout.contents()).not.toContain("runx history [query]"); + expect(stdout.contents()).not.toContain("runx add "); + expect(stdout.contents()).not.toContain("runx mcp serve "); }); it("rejects retired command aliases and TS-only history helpers", async () => { diff --git a/packages/cli/src/native-registry.ts b/packages/cli/src/native-registry.ts index 4e0774049..386e00041 100644 --- a/packages/cli/src/native-registry.ts +++ b/packages/cli/src/native-registry.ts @@ -67,6 +67,8 @@ function normalizeRustRegistrySearchResult(value: unknown): SkillSearchResult { !result || typeof result.skill_id !== "string" || typeof result.name !== "string" || + (result.category !== undefined && typeof result.category !== "string") || + (result.source_category !== undefined && typeof result.source_category !== "string") || typeof result.owner !== "string" || result.source !== "runx-registry" || typeof result.source_label !== "string" || diff --git a/packages/cli/src/official-skills.lock.json b/packages/cli/src/official-skills.lock.json index 6789cb96b..25c504ddc 100644 --- a/packages/cli/src/official-skills.lock.json +++ b/packages/cli/src/official-skills.lock.json @@ -6,6 +6,13 @@ "catalog_visibility": "public", "catalog_role": "context" }, + { + "skill_id": "runx/business-ops", + "version": "sha-ca97efa3deb0", + "digest": "27c05e8e30d8e925c93ecd51e535add38a4ceeabbfc76acf73bb0a578fee3e11", + "catalog_visibility": "public", + "catalog_role": "canonical" + }, { "skill_id": "runx/charge", "version": "sha-0e2f6aef60db", @@ -27,6 +34,13 @@ "catalog_visibility": "public", "catalog_role": "context" }, + { + "skill_id": "runx/dependency-cve-audit", + "version": "sha-e9e461e41ea3", + "digest": "c19ec9fdeb088daab950b7c2e1f3757880de9702e31e40b57e2f65c0c4033348", + "catalog_visibility": "public", + "catalog_role": "canonical" + }, { "skill_id": "runx/design-skill", "version": "sha-0353a69bc33f", @@ -99,7 +113,7 @@ }, { "skill_id": "runx/issue-intake", - "version": "sha-94c248e98a1c", + "version": "sha-15369469618b", "digest": "cc964980fe249ac3633e7b30c664648f0df9406a0254ede9bb0e3cbcdebdd603", "catalog_visibility": "public", "catalog_role": "context" @@ -141,7 +155,7 @@ }, { "skill_id": "runx/least-privilege-auditor", - "version": "sha-b2913b95c069", + "version": "sha-e5c3622556d9", "digest": "244df5dd8eed7900d1987c76060893d3c9cd65f420c5b8c177b19fa4e0b81ac2", "catalog_visibility": "public", "catalog_role": "canonical" @@ -253,8 +267,8 @@ }, { "skill_id": "runx/pr-review-note", - "version": "sha-dd1608925a76", - "digest": "1b9f34f9e7f5355a10babbd154333db4b1b94fa16668583438166b91eec95e0a", + "version": "sha-537dd9fc3c6b", + "digest": "b073ec884f56c9e412d0c1039d5f28f163df0f5530eb0bee922ed4c557955c52", "catalog_visibility": "public", "catalog_role": "context" }, @@ -316,8 +330,8 @@ }, { "skill_id": "runx/review-skill", - "version": "sha-a27dde2ab9bb", - "digest": "09b4a6ec017f9d75536c6db21c60667bd855a20b0b20f53054f63143cbb9d13d", + "version": "sha-622805df5ff3", + "digest": "6fc1b341d55e3c6be8a5f7693dfe3312654b89a14f88fe42e4ffc84a65a9cd09", "catalog_visibility": "public", "catalog_role": "context" }, @@ -330,8 +344,8 @@ }, { "skill_id": "runx/runx-operator", - "version": "sha-da018df211a3", - "digest": "1c8d199a5dd0812a09eb4e785bcd9bfd7af67ec0ef3227149bed1de150a47fff", + "version": "sha-0fed07a0dc00", + "digest": "9de1d9dffb46b6bb14872b66738d5e9b26f271c6f11595c6a685d4c30e539176", "catalog_visibility": "public", "catalog_role": "canonical" }, @@ -372,8 +386,8 @@ }, { "skill_id": "runx/skill-testing", - "version": "sha-c31b54a981c8", - "digest": "7fc86c62bd493cb374850d7e9fc4faad94adb318fc3b20947aa2d411a741cc75", + "version": "sha-9113dacaa62a", + "digest": "93f7a0c009e289862fcc9236effdf0ac75197e9eb042a83200720d23d01cb443", "catalog_visibility": "public", "catalog_role": "context" }, @@ -386,8 +400,8 @@ }, { "skill_id": "runx/sourcey", - "version": "sha-d025d3a4701e", - "digest": "2bdffb5206cbfc2dc619ffead5d26ad192afe0f2836093d782c7901841713006", + "version": "sha-2b08f620e0fa", + "digest": "4b6316c7fbb323b7d27d304deb8f11cb8f939dc31e0b74349d56f27abf618504", "catalog_visibility": "public", "catalog_role": "context" }, @@ -426,6 +440,20 @@ "catalog_visibility": "internal", "catalog_role": "runtime-path" }, + { + "skill_id": "runx/structured-extraction", + "version": "sha-f14902374e11", + "digest": "ebe37921d3b9cb63aa1ca5232a8075607d3b4ad541af4c1bc901ace749ea404f", + "catalog_visibility": "public", + "catalog_role": "canonical" + }, + { + "skill_id": "runx/support-triage-reply", + "version": "sha-fce24eb780f9", + "digest": "94beeed742142c6a236eb0ae1db5644d3f868362030089332c6ce3b40b1f86bb", + "catalog_visibility": "public", + "catalog_role": "canonical" + }, { "skill_id": "runx/taste-profile", "version": "sha-30ae4695f7a2", @@ -470,8 +498,8 @@ }, { "skill_id": "runx/write-harness", - "version": "sha-c989640c5604", - "digest": "f4fbf60192335baff43a5d50f3702a17f96a42a25d69508f457cf0e396320528", + "version": "sha-f69b01f883e0", + "digest": "8fbac78e4c760a124c704ce62aa11ecb8b65b165c72a81fd2c1de163c5bb259b", "catalog_visibility": "public", "catalog_role": "context" }, diff --git a/packages/cli/src/presentation/init-new.ts b/packages/cli/src/presentation/init-new.ts index d642e97f9..1dc7099c3 100644 --- a/packages/cli/src/presentation/init-new.ts +++ b/packages/cli/src/presentation/init-new.ts @@ -8,8 +8,7 @@ export function renderNewResult(result: NewResult, env: NodeJS.ProcessEnv = proc "runx new", "success", [ - ["package", result.name], - ["packet_namespace", result.packet_namespace], + ["skill", result.name], ["directory", result.directory], ["files", String(result.files.length)], ["next", result.next_steps.join(" && ")], diff --git a/packages/cli/src/presentation/search.ts b/packages/cli/src/presentation/search.ts index aa2974664..22bf3b457 100644 --- a/packages/cli/src/presentation/search.ts +++ b/packages/cli/src/presentation/search.ts @@ -62,6 +62,12 @@ export function renderSearchResults( if (result.summary) { lines.push(` ${t.dim}${result.summary}${t.reset}`); } + if (result.category) { + lines.push(` ${t.dim}category:${t.reset} ${result.category}`); + } + if (!result.category && result.source_category) { + lines.push(` ${t.dim}source category:${t.reset} ${result.source_category}`); + } if (result.profile_mode === "profiled" && result.runner_names.length > 0) { lines.push(` ${t.dim}runners:${t.reset} ${result.runner_names.join(", ")}`); } diff --git a/packages/cli/src/scaffold.ts b/packages/cli/src/scaffold.ts index 3b6891743..abe81aebe 100644 --- a/packages/cli/src/scaffold.ts +++ b/packages/cli/src/scaffold.ts @@ -1,16 +1,8 @@ import { mkdir, readdir, writeFile } from "node:fs/promises"; import path from "node:path"; -import { sha256Prefixed } from "@runxhq/contracts"; import { isNodeError } from "./cli-util.js"; -import { sha256Stable } from "./authoring-utils.js"; -import { readCliDependencyVersion, readCliPackageMetadata } from "./metadata.js"; - -const toolkitVersion = readCliDependencyVersion("@runxhq/authoring"); -const authoringPackageVersion = `^${toolkitVersion}`; -const cliPackageVersion = `^${readCliPackageMetadata().version}`; - export interface ScaffoldRunxPackageOptions { readonly name: string; readonly directory: string; @@ -18,7 +10,6 @@ export interface ScaffoldRunxPackageOptions { export interface ScaffoldRunxPackageResult { readonly name: string; - readonly packet_namespace: string; readonly directory: string; readonly files: readonly string[]; readonly next_steps: readonly string[]; @@ -26,315 +17,158 @@ export interface ScaffoldRunxPackageResult { export async function scaffoldRunxPackage(options: ScaffoldRunxPackageOptions): Promise { const name = sanitizeRunxPackageName(options.name); - const packetNamespace = packetNamespaceForName(name); const root = path.resolve(options.directory); await assertWritableScaffoldTarget(root); - const packetId = `${packetNamespace}.echo.v1`; - const toolSource = `import { defineTool, stringInput } from "@runxhq/authoring"; + const writes = scaffoldPackageFiles(name); + await mkdir(root, { recursive: true }); + await Promise.all(writes.map(([relativePath, contents]) => write(root, relativePath, contents))); -export default defineTool({ - name: "docs.echo", - version: "0.1.0", - description: "Echo a docs message.", - inputs: { - message: stringInput({ default: "hello" }), - }, - output: { - packet: "${packetId}", - wrap_as: "echo_packet", - }, - scopes: ["docs.read"], - run({ inputs }) { - return { message: inputs.message }; - }, -}); -`; - const toolRuntime = `const fs = require("node:fs"); -const rawInputs = process.env.RUNX_INPUTS_PATH - ? fs.readFileSync(process.env.RUNX_INPUTS_PATH, "utf8") - : (process.env.RUNX_INPUTS_JSON || "{}"); -const inputs = JSON.parse(rawInputs); -process.stdout.write(JSON.stringify({ schema: "${packetId}", data: { message: String(inputs.message || "hello") } })); -`; - const toolInputs = { - message: { type: "string", required: false, default: "hello" }, - }; - const toolOutput = { packet: packetId, wrap_as: "echo_packet" }; - const toolRunx = { artifacts: { wrap_as: "echo_packet" } }; - const sourceHash = sha256ToolSourceContents({ - "src/index.ts": toolSource, - "run.mjs": toolRuntime, - }); - const schemaHash = sha256Stable({ - inputs: toolInputs, - output: toolOutput, - artifacts: toolRunx.artifacts, - }); - const agentFixture = { - target: { kind: "skill", ref: "." }, - inputs: { message: "hello" }, - agent: { mode: "replay" }, - expect: { - status: "sealed", - outputs: { - echo_packet: { - matches_packet: packetId, - }, - }, - }, + return { + name, + directory: root, + files: writes.map(([relativePath]) => relativePath), + next_steps: [ + `cd ${root}`, + "runx harness . --json", + "runx skill . --input message=hello --json", + ], }; +} - const writes: ReadonlyArray = [ - ["package.json", JSON.stringify({ - name, - version: "0.1.0", - description: "Scaffolded runx skill package.", - type: "module", - publishConfig: { - access: "public", - }, - scripts: { - build: "runx tool build --all --json", - "runx:list": "runx list --json", - "runx:doctor": "runx doctor --json", - "runx:dev": "runx dev --lane deterministic --json", - prepublishOnly: "runx tool build --all --json && runx doctor --json", - }, - runx: { - packets: ["./dist/packets/*.schema.json"], - }, - devDependencies: { - "@runxhq/authoring": authoringPackageVersion, - "@runxhq/cli": cliPackageVersion, - "@tsconfig/node20": "^20.1.6", - "tsx": "^4.20.6", - }, - }, null, 2)], - ["README.md", `# ${name} - -Runx authoring package: composable skills governed by typed contracts. - -## Layout - -- \`SKILL.md\`: Anthropic-standard skill description. Read by humans and agents. -- \`X.yaml\`: runx execution profile layered on top of \`SKILL.md\`. -- \`src/packets/\`: typed packet contracts authored with TypeBox. -- \`tools/\`: deterministic implementation units authored with \`defineTool\`. -- \`fixtures/\`: examples and tests across deterministic, agent, and repo-integration lanes. - -## Authoring Loop - -\`\`\`bash -pnpm install -pnpm build -pnpm runx:list -pnpm runx:doctor -pnpm runx:dev -\`\`\` - -Edit \`tools/docs/echo/src/index.ts\`, then run \`runx tool build --all\` to regenerate \`manifest.json\` and \`run.mjs\`. Add fixtures in \`tools///fixtures/\` to lock behaviour. +export function sanitizeRunxPackageName(value: string): string { + return value.trim().toLowerCase().replace(/[^a-z0-9_.-]+/g, "-").replace(/^[._-]+|[._-]+$/g, "") || "runx-package"; +} -Packet IDs are immutable. Schema changes mean a new packet ID, not an in-place edit. +function scaffoldPackageFiles(name: string): ReadonlyArray { + return [ + ["SKILL.md", skillMd(name)], + ["X.yaml", xYaml(name)], + ["run.mjs", runMjs()], + ["README.md", readme(name)], + [".gitignore", "node_modules/\n.runx/\n*.tgz\n"], + ]; +} -## Bootstrap +function skillMd(name: string): string { + return `--- +name: ${name} +description: ${name} runx skill. Replace this with what the skill does and returns. +source: + type: cli-tool + command: node + args: + - run.mjs + timeout_seconds: 30 + sandbox: + profile: readonly + cwd_policy: skill-directory +inputs: + message: + type: string + required: true + description: Input the skill acts on. Replace with the real inputs. +runx: + category: ops + input_resolution: + required: + - message +--- -- Canonical: \`runx new ${name}\` -- Cold start: \`npm create @runxhq/skill@latest ${name}\` +# ${name} -## Publish +Describe what this skill does, when an agent should reach for it, and what it +returns. Replace the echo in \`run.mjs\` with the real work, and add cases to +\`X.yaml\` so the behaviour is locked by the harness. +`; +} -The scaffold includes \`.github/workflows/publish.yml\`, which publishes with npm provenance from GitHub Actions. Before publishing, update \`package.json\` metadata for your repo and package. -`], - ["SKILL.md", `--- -name: ${name} -description: Scaffolded runx skill package. ---- +function xYaml(name: string): string { + return `skill: ${name} +version: "0.1.0" -Use this skill to demonstrate a governed runx authoring package. -`], - ["X.yaml", `skill: ${name} +catalog: + kind: skill + audience: public + visibility: public + role: canonical + +harness: + cases: + - name: ${name}-smoke + runner: default + inputs: + message: hello + expect: + status: sealed + receipt: + schema: runx.receipt.v1 + state: sealed + disposition: closed + reason_code: process_closed + - name: ${name}-empty-message-fails + runner: default + inputs: + message: "" + expect: + status: failure + receipt: + schema: runx.receipt.v1 + state: sealed + disposition: closed + reason_code: process_failed runners: default: default: true - type: graph + type: cli-tool + command: node + args: + - run.mjs inputs: message: type: string - required: false - default: hello - graph: - name: ${name} - steps: - - id: echo - tool: docs.echo - inputs: - message: inputs.message -`], - ["src/packets/echo.ts", `import { definePacket, t } from "@runxhq/authoring"; + required: true + description: Input the skill acts on. +`; +} -export const EchoPacket = definePacket({ - id: "${packetId}", - schema: t.Object({ - message: t.String(), - }), -}); -`], - ["dist/packets/echo.v1.schema.json", JSON.stringify({ - "$schema": "https://json-schema.org/draft/2020-12/schema", - "$id": `https://schemas.runx.dev/${packetNamespace.replaceAll(".", "/")}/echo/v1.json`, - "x-runx-packet-id": packetId, - type: "object", - required: ["message"], - properties: { - message: { type: "string" }, - }, - additionalProperties: false, - }, null, 2)], - ["tools/docs/echo/src/index.ts", toolSource], - ["tools/docs/echo/run.mjs", toolRuntime], - ["tools/docs/echo/manifest.json", JSON.stringify({ - schema: "runx.tool.manifest.v1", - name: "docs.echo", - version: "0.1.0", - description: "Echo a docs message.", - source: { type: "cli-tool", command: "node", args: ["./run.mjs"] }, - runtime: { command: "node", args: ["./run.mjs"] }, - inputs: toolInputs, - output: toolOutput, - scopes: ["docs.read"], - runx: toolRunx, - source_hash: sourceHash, - schema_hash: schemaHash, - toolkit_version: toolkitVersion, - }, null, 2)], - ["tools/docs/echo/fixtures/basic.yaml", `name: echo-basic -lane: deterministic -target: - kind: tool - ref: docs.echo -inputs: - message: hello -expect: - status: sealed - output: - subset: - schema: ${packetId} - data: - message: hello -`], - ["fixtures/agent.yaml", `name: echo-agent-replay -lane: agent -target: - kind: skill - ref: . -inputs: - message: hello -agent: - mode: replay -expect: - status: sealed - outputs: - echo_packet: - matches_packet: ${packetId} -`], - ["fixtures/agent.replay.json", JSON.stringify({ - schema: "runx.replay.v1", - fixture: "echo-agent-replay", - prompt_fingerprint: sha256Stable(agentFixture), - recorded_at: "1970-01-01T00:00:00.000Z", - target: agentFixture.target, - status: "sealed", - outputs: { - echo_packet: { - schema: packetId, - data: { - message: "hello", - }, - }, - }, - usage: { - mode: "scaffold", - }, - }, null, 2)], - ["fixtures/repos/readme-only/README.md", `# ${name} -`], - [".github/workflows/publish.yml", `name: publish +function runMjs(): string { + return `// Inputs arrive as RUNX_INPUT_ environment variables. Do the work and +// write the result to stdout. Replace this echo with the real logic. +const message = process.env.RUNX_INPUT_MESSAGE ?? ""; +if (message.trim().length === 0) { + process.stderr.write("message is required\\n"); + process.exit(64); +} +process.stdout.write(\`${"${message}"}\\n\`); +`; +} -on: - workflow_dispatch: - release: - types: - - published +function readme(name: string): string { + return `# ${name} -jobs: - publish: - runs-on: ubuntu-latest - permissions: - contents: read - id-token: write - steps: - - uses: actions/checkout@v4 - - uses: pnpm/action-setup@v4 - with: - version: 10 - - uses: actions/setup-node@v4 - with: - node-version: 20 - registry-url: https://registry.npmjs.org - cache: pnpm - - run: pnpm install --frozen-lockfile - - run: pnpm build - - run: pnpm runx:doctor - - run: npm publish --provenance --access public - env: - NODE_AUTH_TOKEN: \${{ secrets.NPM_TOKEN }} -`], - [".gitignore", `node_modules/ -.runx/ -*.tgz -`], - [".gitattributes", "tools/**/run.mjs linguist-generated=true\ntools/**/manifest.json linguist-generated=true\ntools/**/dist/** linguist-generated=true\n"], - ["tsconfig.json", JSON.stringify({ - extends: "@tsconfig/node20/tsconfig.json", - compilerOptions: { - module: "NodeNext", - moduleResolution: "NodeNext", - strict: true, - }, - include: ["src/**/*.ts", "tools/**/*.ts"], - }, null, 2)], - ]; +A native runx skill: a \`SKILL.md\` contract, an \`X.yaml\` execution profile, and a +\`run.mjs\` script. No build step and no dependencies. - await mkdir(root, { recursive: true }); - await Promise.all(writes.map(([relativePath, contents]) => write(root, relativePath, contents))); +## Develop - return { - name, - packet_namespace: packetNamespace, - directory: root, - files: writes.map(([relativePath]) => relativePath), - next_steps: [ - `cd ${root}`, - "pnpm install", - "pnpm build", - "runx dev", - ], - }; -} +\`\`\`bash +runx harness . --json # run the harness cases in X.yaml +runx skill . --input message=hello --json # run the skill once +runx history # inspect the signed receipt +\`\`\` -export function sanitizeRunxPackageName(value: string): string { - return value.trim().toLowerCase().replace(/[^a-z0-9_.-]+/g, "-").replace(/^[._-]+|[._-]+$/g, "") || "runx-package"; -} +Edit \`run.mjs\` to do the real work, and keep both harness classes in \`X.yaml\`: +one happy path and one stop, error, or refusal case. -function packetNamespaceForName(value: string): string { - return value - .toLowerCase() - .replace(/^@/, "") - .replace(/[^a-z0-9]+/g, ".") - .replace(/^\.+|\.+$/g, "") - || "runx.package"; +## Publish + +\`\`\`bash +runx login --provider github --for publish +runx registry publish . # the registry runs the harness as the publish gate +\`\`\` +`; } async function assertWritableScaffoldTarget(root: string): Promise { @@ -354,19 +188,3 @@ async function write(root: string, relativePath: string, contents: string): Prom await mkdir(path.dirname(filePath), { recursive: true }); await writeFile(filePath, contents.endsWith("\n") ? contents : `${contents}\n`); } - -function sha256ToolSourceContents(files: Readonly>): string { - const chunks: Uint8Array[] = []; - for (const relativePath of ["src/index.ts", "run.mjs"]) { - if (files[relativePath] === undefined) { - continue; - } - chunks.push( - Buffer.from(relativePath), - Buffer.from("\0"), - Buffer.from(files[relativePath] ?? ""), - Buffer.from("\0"), - ); - } - return sha256Prefixed(Buffer.concat(chunks)); -} diff --git a/packages/cli/src/skill-refs.test.ts b/packages/cli/src/skill-refs.test.ts index 98f443491..ca90aa16e 100644 --- a/packages/cli/src/skill-refs.test.ts +++ b/packages/cli/src/skill-refs.test.ts @@ -11,9 +11,11 @@ import { officialSkillVisibleForCatalog } from "./skill-refs.js"; const repoRoot = path.resolve(path.dirname(fileURLToPath(import.meta.url)), "../../.."); const publicOfficialCatalogSkills = [ "brand-voice", + "business-ops", "charge", "content-pipeline", "deep-research-brief", + "dependency-cve-audit", "design-skill", "dispute-respond", "draft-content", @@ -62,6 +64,8 @@ const publicOfficialCatalogSkills = [ "spend", "sql-analyst", "stripe-pay", + "structured-extraction", + "support-triage-reply", "taste-profile", "vault-unseal", "vuln-scan", diff --git a/packages/cli/src/skill-refs.ts b/packages/cli/src/skill-refs.ts index 21358976f..d39d03817 100644 --- a/packages/cli/src/skill-refs.ts +++ b/packages/cli/src/skill-refs.ts @@ -2,6 +2,7 @@ import { existsSync, readFileSync } from "node:fs"; import { cp, mkdir, readFile, readdir, rm, writeFile } from "node:fs/promises"; import path from "node:path"; import { fileURLToPath } from "node:url"; +import { parseDocument } from "yaml"; import { resolvePathFromUserInput, @@ -341,9 +342,9 @@ async function searchBundledSkills(query: string, env: NodeJS.ProcessEnv): Promi const skillMdPath = path.join(bundledDir, entry.name, "SKILL.md"); if (!existsSync(skillMdPath)) continue; const raw = await readFile(skillMdPath, "utf8"); - const { name, description } = parseSkillFrontmatter(raw, entry.name); + const { name, description, category, sourceCategory } = parseSkillFrontmatter(raw, entry.name); if (!officialSkillVisibleForCatalog(`runx/${name}`, env)) continue; - const hay = `${name}\n${description}`.toLowerCase(); + const hay = `${name}\n${description}\n${category ?? ""}\n${sourceCategory ?? ""}`.toLowerCase(); if (needle && !hay.includes(needle)) continue; const hasProfile = existsSync(path.join(path.dirname(bundledDir), "bindings", "runx", entry.name, "X.yaml")); out.push({ @@ -356,7 +357,9 @@ async function searchBundledSkills(query: string, env: NodeJS.ProcessEnv): Promi source_type: "bundled", trust_tier: "first_party", required_scopes: [], - tags: [], + tags: category ? [category] : [], + category, + source_category: sourceCategory, profile_mode: hasProfile ? "profiled" : "portable", runner_names: [], add_command: `runx add runx/${name}`, @@ -483,18 +486,44 @@ function assertSkillReferencePath(resolved: string): void { } } -function parseSkillFrontmatter(raw: string, fallbackName: string): { name: string; description: string } { +function parseSkillFrontmatter(raw: string, fallbackName: string): { name: string; description: string; category?: string; sourceCategory?: string } { const match = raw.match(/^---\n([\s\S]*?)\n---/); let name = fallbackName; let description = ""; + let category: string | undefined; + let sourceCategory: string | undefined; if (match) { + const parsed = parseDocument(match[1], { prettyErrors: false }); + if (parsed.errors.length === 0) { + const frontmatter = asRecord(parsed.toJS()); + if (frontmatter) { + const runx = asRecord(frontmatter.runx); + const runxCategory = normalizeCategory(stringValue(runx?.category)); + return { + name: stringValue(frontmatter.name) || fallbackName, + description: stringValue(frontmatter.description) ?? "", + category: runxCategory, + sourceCategory: normalizeCategory(stringValue(frontmatter.category)), + }; + } + } for (const line of match[1].split("\n")) { - const kv = line.match(/^(name|description):\s*(.*)$/); + const kv = line.match(/^(name|description|category):\s*(.*)$/); if (!kv) continue; const value = kv[2].trim().replace(/^["']|["']$/g, ""); if (kv[1] === "name") name = value || fallbackName; else if (kv[1] === "description") description = value; + else if (kv[1] === "category") sourceCategory = normalizeCategory(value); } } - return { name, description }; + return { name, description, category, sourceCategory }; +} + +function stringValue(value: unknown): string | undefined { + return typeof value === "string" ? value : undefined; +} + +function normalizeCategory(value: string | undefined): string | undefined { + const normalized = value?.trim(); + return normalized ? normalized : undefined; } diff --git a/packages/contracts/src/index.ts b/packages/contracts/src/index.ts index 99e5254c7..64e86e37b 100644 --- a/packages/contracts/src/index.ts +++ b/packages/contracts/src/index.ts @@ -279,7 +279,6 @@ export { type OperationalProposalRecommendedActionContract, type OperationalProposalReferenceContract, type OperationalProposalReferenceLinkContract, - type OperationalProposalReferenceTypeContract, type OperationalProposalRedactionStatusContract, type OperationalProposalEscalationExtensionContract, type OperationalProposalExtensionsContract, diff --git a/packages/contracts/src/schema-artifacts.ts b/packages/contracts/src/schema-artifacts.ts index 432f894c6..41770a2b8 100644 --- a/packages/contracts/src/schema-artifacts.ts +++ b/packages/contracts/src/schema-artifacts.ts @@ -25028,7 +25028,7 @@ export const runxSchemaArtifacts = { }, "artifact_refs": { "items": { - "$id": "https://schemas.runx.dev/runx/operational-proposal/reference/v1.json", + "$id": "https://schemas.runx.dev/runx/reference/v1.json", "$schema": "https://json-schema.org/draft/2020-12/schema", "additionalProperties": false, "properties": { @@ -25066,11 +25066,31 @@ export const runxSchemaArtifacts = { "type": "string" }, "schema": { - "const": "runx.operational_proposal.reference.v1", + "const": "runx.reference.v1", "type": "string" }, "type": { "anyOf": [ + { + "const": "github_issue", + "type": "string" + }, + { + "const": "github_pull_request", + "type": "string" + }, + { + "const": "github_repo", + "type": "string" + }, + { + "const": "slack_thread", + "type": "string" + }, + { + "const": "sentry_event", + "type": "string" + }, { "const": "provider_thread", "type": "string" @@ -25227,7 +25247,7 @@ export const runxSchemaArtifacts = { "uri" ], "type": "object", - "x-runx-schema": "runx.operational_proposal.reference.v1" + "x-runx-schema": "runx.reference.v1" }, "type": "array" }, @@ -25284,7 +25304,7 @@ export const runxSchemaArtifacts = { }, "evidence_refs": { "items": { - "$id": "https://schemas.runx.dev/runx/operational-proposal/reference/v1.json", + "$id": "https://schemas.runx.dev/runx/reference/v1.json", "$schema": "https://json-schema.org/draft/2020-12/schema", "additionalProperties": false, "properties": { @@ -25322,11 +25342,31 @@ export const runxSchemaArtifacts = { "type": "string" }, "schema": { - "const": "runx.operational_proposal.reference.v1", + "const": "runx.reference.v1", "type": "string" }, "type": { "anyOf": [ + { + "const": "github_issue", + "type": "string" + }, + { + "const": "github_pull_request", + "type": "string" + }, + { + "const": "github_repo", + "type": "string" + }, + { + "const": "slack_thread", + "type": "string" + }, + { + "const": "sentry_event", + "type": "string" + }, { "const": "provider_thread", "type": "string" @@ -25483,7 +25523,7 @@ export const runxSchemaArtifacts = { "uri" ], "type": "object", - "x-runx-schema": "runx.operational_proposal.reference.v1" + "x-runx-schema": "runx.reference.v1" }, "type": "array" }, @@ -25504,7 +25544,7 @@ export const runxSchemaArtifacts = { }, "refs": { "items": { - "$id": "https://schemas.runx.dev/runx/operational-proposal/reference/v1.json", + "$id": "https://schemas.runx.dev/runx/reference/v1.json", "$schema": "https://json-schema.org/draft/2020-12/schema", "additionalProperties": false, "properties": { @@ -25542,11 +25582,31 @@ export const runxSchemaArtifacts = { "type": "string" }, "schema": { - "const": "runx.operational_proposal.reference.v1", + "const": "runx.reference.v1", "type": "string" }, "type": { "anyOf": [ + { + "const": "github_issue", + "type": "string" + }, + { + "const": "github_pull_request", + "type": "string" + }, + { + "const": "github_repo", + "type": "string" + }, + { + "const": "slack_thread", + "type": "string" + }, + { + "const": "sentry_event", + "type": "string" + }, { "const": "provider_thread", "type": "string" @@ -25703,7 +25763,7 @@ export const runxSchemaArtifacts = { "uri" ], "type": "object", - "x-runx-schema": "runx.operational_proposal.reference.v1" + "x-runx-schema": "runx.reference.v1" }, "type": "array" }, @@ -25759,7 +25819,7 @@ export const runxSchemaArtifacts = { "type": "array" }, "hydrated_context_ref": { - "$id": "https://schemas.runx.dev/runx/operational-proposal/reference/v1.json", + "$id": "https://schemas.runx.dev/runx/reference/v1.json", "$schema": "https://json-schema.org/draft/2020-12/schema", "additionalProperties": false, "properties": { @@ -25797,11 +25857,31 @@ export const runxSchemaArtifacts = { "type": "string" }, "schema": { - "const": "runx.operational_proposal.reference.v1", + "const": "runx.reference.v1", "type": "string" }, "type": { "anyOf": [ + { + "const": "github_issue", + "type": "string" + }, + { + "const": "github_pull_request", + "type": "string" + }, + { + "const": "github_repo", + "type": "string" + }, + { + "const": "slack_thread", + "type": "string" + }, + { + "const": "sentry_event", + "type": "string" + }, { "const": "provider_thread", "type": "string" @@ -25958,7 +26038,7 @@ export const runxSchemaArtifacts = { "uri" ], "type": "object", - "x-runx-schema": "runx.operational_proposal.reference.v1" + "x-runx-schema": "runx.reference.v1" }, "idempotency": { "additionalProperties": false, @@ -26005,12 +26085,12 @@ export const runxSchemaArtifacts = { }, "publication_refs": { "items": { - "$id": "https://schemas.runx.dev/runx/operational-proposal/reference-link/v1.json", + "$id": "https://schemas.runx.dev/runx/reference-link/v1.json", "$schema": "https://json-schema.org/draft/2020-12/schema", "additionalProperties": false, "properties": { "ref": { - "$id": "https://schemas.runx.dev/runx/operational-proposal/reference/v1.json", + "$id": "https://schemas.runx.dev/runx/reference/v1.json", "$schema": "https://json-schema.org/draft/2020-12/schema", "additionalProperties": false, "properties": { @@ -26048,11 +26128,31 @@ export const runxSchemaArtifacts = { "type": "string" }, "schema": { - "const": "runx.operational_proposal.reference.v1", + "const": "runx.reference.v1", "type": "string" }, "type": { "anyOf": [ + { + "const": "github_issue", + "type": "string" + }, + { + "const": "github_pull_request", + "type": "string" + }, + { + "const": "github_repo", + "type": "string" + }, + { + "const": "slack_thread", + "type": "string" + }, + { + "const": "sentry_event", + "type": "string" + }, { "const": "provider_thread", "type": "string" @@ -26209,14 +26309,14 @@ export const runxSchemaArtifacts = { "uri" ], "type": "object", - "x-runx-schema": "runx.operational_proposal.reference.v1" + "x-runx-schema": "runx.reference.v1" }, "role": { "minLength": 1, "type": "string" }, "schema": { - "const": "runx.operational_proposal.reference_link.v1", + "const": "runx.reference_link.v1", "type": "string" } }, @@ -26225,7 +26325,7 @@ export const runxSchemaArtifacts = { "ref" ], "type": "object", - "x-runx-schema": "runx.operational_proposal.reference_link.v1" + "x-runx-schema": "runx.reference_link.v1" }, "type": "array" }, @@ -26235,7 +26335,7 @@ export const runxSchemaArtifacts = { }, "receipt_refs": { "items": { - "$id": "https://schemas.runx.dev/runx/operational-proposal/reference/v1.json", + "$id": "https://schemas.runx.dev/runx/reference/v1.json", "$schema": "https://json-schema.org/draft/2020-12/schema", "additionalProperties": false, "properties": { @@ -26273,11 +26373,31 @@ export const runxSchemaArtifacts = { "type": "string" }, "schema": { - "const": "runx.operational_proposal.reference.v1", + "const": "runx.reference.v1", "type": "string" }, "type": { "anyOf": [ + { + "const": "github_issue", + "type": "string" + }, + { + "const": "github_pull_request", + "type": "string" + }, + { + "const": "github_repo", + "type": "string" + }, + { + "const": "slack_thread", + "type": "string" + }, + { + "const": "sentry_event", + "type": "string" + }, { "const": "provider_thread", "type": "string" @@ -26434,7 +26554,7 @@ export const runxSchemaArtifacts = { "uri" ], "type": "object", - "x-runx-schema": "runx.operational_proposal.reference.v1" + "x-runx-schema": "runx.reference.v1" }, "type": "array" }, @@ -26455,7 +26575,7 @@ export const runxSchemaArtifacts = { }, "target_refs": { "items": { - "$id": "https://schemas.runx.dev/runx/operational-proposal/reference/v1.json", + "$id": "https://schemas.runx.dev/runx/reference/v1.json", "$schema": "https://json-schema.org/draft/2020-12/schema", "additionalProperties": false, "properties": { @@ -26493,11 +26613,31 @@ export const runxSchemaArtifacts = { "type": "string" }, "schema": { - "const": "runx.operational_proposal.reference.v1", + "const": "runx.reference.v1", "type": "string" }, "type": { "anyOf": [ + { + "const": "github_issue", + "type": "string" + }, + { + "const": "github_pull_request", + "type": "string" + }, + { + "const": "github_repo", + "type": "string" + }, + { + "const": "slack_thread", + "type": "string" + }, + { + "const": "sentry_event", + "type": "string" + }, { "const": "provider_thread", "type": "string" @@ -26654,7 +26794,7 @@ export const runxSchemaArtifacts = { "uri" ], "type": "object", - "x-runx-schema": "runx.operational_proposal.reference.v1" + "x-runx-schema": "runx.reference.v1" }, "type": "array" } @@ -26686,12 +26826,12 @@ export const runxSchemaArtifacts = { }, "result_refs": { "items": { - "$id": "https://schemas.runx.dev/runx/operational-proposal/reference-link/v1.json", + "$id": "https://schemas.runx.dev/runx/reference-link/v1.json", "$schema": "https://json-schema.org/draft/2020-12/schema", "additionalProperties": false, "properties": { "ref": { - "$id": "https://schemas.runx.dev/runx/operational-proposal/reference/v1.json", + "$id": "https://schemas.runx.dev/runx/reference/v1.json", "$schema": "https://json-schema.org/draft/2020-12/schema", "additionalProperties": false, "properties": { @@ -26729,11 +26869,31 @@ export const runxSchemaArtifacts = { "type": "string" }, "schema": { - "const": "runx.operational_proposal.reference.v1", + "const": "runx.reference.v1", "type": "string" }, "type": { "anyOf": [ + { + "const": "github_issue", + "type": "string" + }, + { + "const": "github_pull_request", + "type": "string" + }, + { + "const": "github_repo", + "type": "string" + }, + { + "const": "slack_thread", + "type": "string" + }, + { + "const": "sentry_event", + "type": "string" + }, { "const": "provider_thread", "type": "string" @@ -26890,14 +27050,14 @@ export const runxSchemaArtifacts = { "uri" ], "type": "object", - "x-runx-schema": "runx.operational_proposal.reference.v1" + "x-runx-schema": "runx.reference.v1" }, "role": { "minLength": 1, "type": "string" }, "schema": { - "const": "runx.operational_proposal.reference_link.v1", + "const": "runx.reference_link.v1", "type": "string" } }, @@ -26906,7 +27066,7 @@ export const runxSchemaArtifacts = { "ref" ], "type": "object", - "x-runx-schema": "runx.operational_proposal.reference_link.v1" + "x-runx-schema": "runx.reference_link.v1" }, "type": "array" }, @@ -26931,7 +27091,7 @@ export const runxSchemaArtifacts = { "type": "string" }, "source_ref": { - "$id": "https://schemas.runx.dev/runx/operational-proposal/reference/v1.json", + "$id": "https://schemas.runx.dev/runx/reference/v1.json", "$schema": "https://json-schema.org/draft/2020-12/schema", "additionalProperties": false, "properties": { @@ -26969,11 +27129,31 @@ export const runxSchemaArtifacts = { "type": "string" }, "schema": { - "const": "runx.operational_proposal.reference.v1", + "const": "runx.reference.v1", "type": "string" }, "type": { "anyOf": [ + { + "const": "github_issue", + "type": "string" + }, + { + "const": "github_pull_request", + "type": "string" + }, + { + "const": "github_repo", + "type": "string" + }, + { + "const": "slack_thread", + "type": "string" + }, + { + "const": "sentry_event", + "type": "string" + }, { "const": "provider_thread", "type": "string" @@ -27130,10 +27310,10 @@ export const runxSchemaArtifacts = { "uri" ], "type": "object", - "x-runx-schema": "runx.operational_proposal.reference.v1" + "x-runx-schema": "runx.reference.v1" }, "source_thread_ref": { - "$id": "https://schemas.runx.dev/runx/operational-proposal/reference/v1.json", + "$id": "https://schemas.runx.dev/runx/reference/v1.json", "$schema": "https://json-schema.org/draft/2020-12/schema", "additionalProperties": false, "properties": { @@ -27171,11 +27351,31 @@ export const runxSchemaArtifacts = { "type": "string" }, "schema": { - "const": "runx.operational_proposal.reference.v1", + "const": "runx.reference.v1", "type": "string" }, "type": { "anyOf": [ + { + "const": "github_issue", + "type": "string" + }, + { + "const": "github_pull_request", + "type": "string" + }, + { + "const": "github_repo", + "type": "string" + }, + { + "const": "slack_thread", + "type": "string" + }, + { + "const": "sentry_event", + "type": "string" + }, { "const": "provider_thread", "type": "string" @@ -27332,11 +27532,11 @@ export const runxSchemaArtifacts = { "uri" ], "type": "object", - "x-runx-schema": "runx.operational_proposal.reference.v1" + "x-runx-schema": "runx.reference.v1" }, "story_refs": { "items": { - "$id": "https://schemas.runx.dev/runx/operational-proposal/reference/v1.json", + "$id": "https://schemas.runx.dev/runx/reference/v1.json", "$schema": "https://json-schema.org/draft/2020-12/schema", "additionalProperties": false, "properties": { @@ -27374,11 +27574,31 @@ export const runxSchemaArtifacts = { "type": "string" }, "schema": { - "const": "runx.operational_proposal.reference.v1", + "const": "runx.reference.v1", "type": "string" }, "type": { "anyOf": [ + { + "const": "github_issue", + "type": "string" + }, + { + "const": "github_pull_request", + "type": "string" + }, + { + "const": "github_repo", + "type": "string" + }, + { + "const": "slack_thread", + "type": "string" + }, + { + "const": "sentry_event", + "type": "string" + }, { "const": "provider_thread", "type": "string" @@ -27535,7 +27755,7 @@ export const runxSchemaArtifacts = { "uri" ], "type": "object", - "x-runx-schema": "runx.operational_proposal.reference.v1" + "x-runx-schema": "runx.reference.v1" }, "type": "array" } diff --git a/packages/contracts/src/schemas/operational-proposal.test.ts b/packages/contracts/src/schemas/operational-proposal.test.ts index fe1da6506..a3df9d107 100644 --- a/packages/contracts/src/schemas/operational-proposal.test.ts +++ b/packages/contracts/src/schemas/operational-proposal.test.ts @@ -39,7 +39,6 @@ describe("operational proposal schema", () => { "invalid-missing-source-ref.json", "invalid-provider-specific-field.json", "invalid-product-specific-field.json", - "invalid-provider-locked-reference-type.json", ])("rejects invalid fixture %s", (fixtureName) => { expect(() => validateOperationalProposalContract(readExpected(fixtureName))).toThrow(); }); diff --git a/packages/contracts/src/schemas/operational-proposal.ts b/packages/contracts/src/schemas/operational-proposal.ts index f33762c02..87eaaa4b1 100644 --- a/packages/contracts/src/schemas/operational-proposal.ts +++ b/packages/contracts/src/schemas/operational-proposal.ts @@ -3,7 +3,7 @@ import { generatedSchema, validateContractSchema, } from "../internal.js"; -import type { ProofKindContract } from "./spine.js"; +import type { ReferenceContract, ReferenceLinkContract } from "./spine.js"; export const operationalProposalSchemaVersion = "runx.operational_proposal.v1" as const; @@ -48,59 +48,8 @@ export type OperationalProposalOutcomeContract = DeepReadonly<{ refs?: readonly OperationalProposalReferenceContract[]; }>; -export type OperationalProposalReferenceTypeContract = - | "provider_thread" - | "provider_event" - | "provider_comment" - | "tracking_item" - | "change_request" - | "repository" - | "support_ticket" - | "signal" - | "act" - | "receipt" - | "graph_receipt" - | "artifact" - | "verification" - | "harness" - | "host" - | "deployment" - | "surface" - | "target" - | "opportunity" - | "thesis_assessment" - | "selection" - | "skill_binding" - | "target_transition_entry" - | "selection_cycle" - | "decision" - | "reflection_entry" - | "feed_entry" - | "principal" - | "authority_proof" - | "scope_admission" - | "grant" - | "mandate" - | "credential" - | "webhook_delivery" - | "redaction_policy" - | "external_url"; - -export type OperationalProposalReferenceContract = DeepReadonly<{ - schema?: string; - type: OperationalProposalReferenceTypeContract; - uri: string; - provider?: string; - locator?: string; - label?: string; - observed_at?: string; - proof_kind?: ProofKindContract; -}>; - -export type OperationalProposalReferenceLinkContract = DeepReadonly<{ - role: string; - ref: OperationalProposalReferenceContract; -}>; +export type OperationalProposalReferenceContract = ReferenceContract; +export type OperationalProposalReferenceLinkContract = ReferenceLinkContract; export type OperationalProposalEscalationExtensionContract = DeepReadonly<{ severity: string; diff --git a/packages/create-skill/README.md b/packages/create-skill/README.md deleted file mode 100644 index 1406dce37..000000000 --- a/packages/create-skill/README.md +++ /dev/null @@ -1,25 +0,0 @@ -# @runxhq/create-skill - -Initializer package behind: - -```bash -npm create @runxhq/skill@latest my-skill -``` - -The canonical runx command remains: - -```bash -runx new my-skill -``` - -This package is intentionally thin. It invokes the `runx` binary from -`@runxhq/cli` so the scaffolding logic stays in one native CLI path. - -## Rust takeover boundary - -`@runxhq/create-skill` remains a thin npm bootstrapper. After the Rust CLI -cutover it continues to wrap `runx new` through the bundled CLI rather than -reimplementing scaffolding logic. - -See the [TypeScript interop boundary](../../docs/ts-interop-boundary.md) for -the package disposition and ownership rules. diff --git a/packages/create-skill/bin/create-skill.js b/packages/create-skill/bin/create-skill.js deleted file mode 100755 index 26c62c505..000000000 --- a/packages/create-skill/bin/create-skill.js +++ /dev/null @@ -1,27 +0,0 @@ -#!/usr/bin/env node - -import { existsSync } from "node:fs"; -import path from "node:path"; -import { fileURLToPath, pathToFileURL } from "node:url"; -import process from "node:process"; - -const packageRoot = path.resolve(path.dirname(fileURLToPath(import.meta.url)), ".."); -const distEntry = path.join(packageRoot, "dist", "index.js"); - -if (existsSync(distEntry)) { - const { runCreateSkill } = await import(pathToFileURL(distEntry).href); - const exitCode = await runCreateSkill(process.argv.slice(2), { - stdin: process.stdin, - stdout: process.stdout, - stderr: process.stderr, - }); - process.exitCode = exitCode; -} else { - const hint = [ - "create-skill: packaged dist is missing.", - "Run the workspace build before invoking this package.", - `Expected entry: ${distEntry}`, - ].join("\n"); - process.stderr.write(`${hint}\n`); - process.exitCode = 1; -} diff --git a/packages/create-skill/package.json b/packages/create-skill/package.json deleted file mode 100644 index 828f1a358..000000000 --- a/packages/create-skill/package.json +++ /dev/null @@ -1,37 +0,0 @@ -{ - "name": "@runxhq/create-skill", - "version": "0.2.0", - "description": "Cold-start scaffolder for runx standalone skill packages.", - "private": false, - "license": "MIT", - "type": "module", - "homepage": "https://github.com/runxhq/runx", - "bugs": { - "url": "https://github.com/runxhq/runx/issues" - }, - "repository": { - "type": "git", - "url": "git+https://github.com/runxhq/runx.git", - "directory": "packages/create-skill" - }, - "publishConfig": { - "access": "public" - }, - "bin": { - "create-skill": "./bin/create-skill.js" - }, - "files": [ - "README.md", - "bin", - "dist" - ], - "exports": { - ".": { - "types": "./dist/index.d.ts", - "import": "./dist/index.js" - } - }, - "dependencies": { - "@runxhq/cli": "workspace:^0.5.22" - } -} diff --git a/packages/create-skill/src/index.test.ts b/packages/create-skill/src/index.test.ts deleted file mode 100644 index 1328a6f4f..000000000 --- a/packages/create-skill/src/index.test.ts +++ /dev/null @@ -1,81 +0,0 @@ -import { Writable } from "node:stream"; - -import { describe, expect, it } from "vitest"; - -import { runCreateSkill, type CliIo } from "./index.js"; - -class MemoryWritable extends Writable { - #chunks: string[] = []; - - _write(chunk: Buffer | string, _encoding: BufferEncoding, callback: (error?: Error | null) => void): void { - this.#chunks.push(typeof chunk === "string" ? chunk : chunk.toString("utf8")); - callback(); - } - - contents(): string { - return this.#chunks.join(""); - } -} - -function createIo() { - const stdout = new MemoryWritable(); - const stderr = new MemoryWritable(); - return { - stdout, - stderr, - io: { - stdin: process.stdin, - stdout: stdout as unknown as NodeJS.WriteStream, - stderr: stderr as unknown as NodeJS.WriteStream, - } satisfies CliIo, - }; -} - -describe("@runxhq/create-skill", () => { - it("forwards to runx new with the original args", async () => { - const calls: unknown[] = []; - const { io, stdout, stderr } = createIo(); - const env = { ...process.env, INIT_CWD: "/tmp/project-root" }; - - const exitCode = await runCreateSkill( - ["demo-skill", "--directory", "packages/demo-skill"], - io, - env, - async (argv, receivedIo, receivedEnv) => { - calls.push({ argv, receivedIo, receivedEnv }); - return 0; - }, - ); - - expect(exitCode).toBe(0); - expect(stdout.contents()).toBe(""); - expect(stderr.contents()).toBe(""); - expect(calls).toEqual([ - { - argv: ["demo-skill", "--directory", "packages/demo-skill"], - receivedIo: io, - receivedEnv: env, - }, - ]); - }); - - it("prints help to stdout", async () => { - const { io, stdout, stderr } = createIo(); - - const exitCode = await runCreateSkill(["--help"], io, process.env, async () => 1); - - expect(exitCode).toBe(0); - expect(stdout.contents()).toContain("npm create @runxhq/skill@latest "); - expect(stderr.contents()).toBe(""); - }); - - it("requires a package name", async () => { - const { io, stdout, stderr } = createIo(); - - const exitCode = await runCreateSkill([], io, process.env, async () => 0); - - expect(exitCode).toBe(64); - expect(stdout.contents()).toBe(""); - expect(stderr.contents()).toContain("runx new "); - }); -}); diff --git a/packages/create-skill/src/index.ts b/packages/create-skill/src/index.ts deleted file mode 100644 index 199f6123f..000000000 --- a/packages/create-skill/src/index.ts +++ /dev/null @@ -1,82 +0,0 @@ -import { spawn } from "node:child_process"; -import type { Readable, Writable } from "node:stream"; - -export interface CliIo { - readonly stdin: Readable; - readonly stdout: Writable; - readonly stderr: Writable; -} - -export type RunCliLike = ( - argv: readonly string[], - io?: CliIo, - env?: NodeJS.ProcessEnv, -) => Promise; - -const usageLines = [ - "Usage:", - " npm create @runxhq/skill@latest [-- --directory dir]", - " runx new [--directory dir]", - "", - "Notes:", - " runx new is the canonical command.", - " The create package is a cold-start entrypoint for the same scaffolder.", -]; - -export async function runRunxNew( - argv: readonly string[], - io: CliIo = { stdin: process.stdin, stdout: process.stdout, stderr: process.stderr }, - env: NodeJS.ProcessEnv = process.env, -): Promise { - const runxBin = env.RUNX_BIN ?? "runx"; - return await new Promise((resolve) => { - let settled = false; - const finish = (code: number) => { - if (!settled) { - settled = true; - resolve(code); - } - }; - const child = spawn(runxBin, ["new", ...argv], { - env, - stdio: "inherit", - }); - child.on("error", (error) => { - io.stderr.write(`create-skill: failed to start runx: ${error.message}\n`); - finish(127); - }); - child.on("exit", (code, signal) => { - if (signal) { - io.stderr.write(`create-skill: runx exited from signal ${signal}\n`); - finish(1); - return; - } - finish(code ?? 1); - }); - }); -} - -export function writeCreateSkillUsage(stream: Writable): void { - stream.write(`${usageLines.join("\n")}\n`); -} - -export async function runCreateSkill( - argv: readonly string[] = process.argv.slice(2), - io: CliIo = { stdin: process.stdin, stdout: process.stdout, stderr: process.stderr }, - env: NodeJS.ProcessEnv = process.env, - runCliImpl: RunCliLike = runRunxNew, -): Promise { - if (argv.length === 1 && (argv[0] === "--help" || argv[0] === "-h")) { - writeCreateSkillUsage(io.stdout); - return 0; - } - if (argv.length === 0) { - writeCreateSkillUsage(io.stderr); - return 64; - } - return await runCliImpl(argv, io, env); -} - -if (import.meta.url === new URL(process.argv[1] ?? "", "file:").href) { - process.exitCode = await runCreateSkill(); -} diff --git a/pnpm-lock.yaml b/pnpm-lock.yaml index 66463cca0..a8e1403ab 100644 --- a/pnpm-lock.yaml +++ b/pnpm-lock.yaml @@ -50,12 +50,6 @@ importers: specifier: ^8.20.0 version: 8.20.0 - packages/create-skill: - dependencies: - '@runxhq/cli': - specifier: workspace:^0.5.22 - version: link:../cli - packages/host-adapters: dependencies: '@runxhq/contracts': diff --git a/schemas/operational-proposal.schema.json b/schemas/operational-proposal.schema.json index 8f8470b4b..9ca8fd1d3 100644 --- a/schemas/operational-proposal.schema.json +++ b/schemas/operational-proposal.schema.json @@ -12,7 +12,7 @@ }, "artifact_refs": { "items": { - "$id": "https://schemas.runx.dev/runx/operational-proposal/reference/v1.json", + "$id": "https://schemas.runx.dev/runx/reference/v1.json", "$schema": "https://json-schema.org/draft/2020-12/schema", "additionalProperties": false, "properties": { @@ -50,11 +50,31 @@ "type": "string" }, "schema": { - "const": "runx.operational_proposal.reference.v1", + "const": "runx.reference.v1", "type": "string" }, "type": { "anyOf": [ + { + "const": "github_issue", + "type": "string" + }, + { + "const": "github_pull_request", + "type": "string" + }, + { + "const": "github_repo", + "type": "string" + }, + { + "const": "slack_thread", + "type": "string" + }, + { + "const": "sentry_event", + "type": "string" + }, { "const": "provider_thread", "type": "string" @@ -211,7 +231,7 @@ "uri" ], "type": "object", - "x-runx-schema": "runx.operational_proposal.reference.v1" + "x-runx-schema": "runx.reference.v1" }, "type": "array" }, @@ -268,7 +288,7 @@ }, "evidence_refs": { "items": { - "$id": "https://schemas.runx.dev/runx/operational-proposal/reference/v1.json", + "$id": "https://schemas.runx.dev/runx/reference/v1.json", "$schema": "https://json-schema.org/draft/2020-12/schema", "additionalProperties": false, "properties": { @@ -306,11 +326,31 @@ "type": "string" }, "schema": { - "const": "runx.operational_proposal.reference.v1", + "const": "runx.reference.v1", "type": "string" }, "type": { "anyOf": [ + { + "const": "github_issue", + "type": "string" + }, + { + "const": "github_pull_request", + "type": "string" + }, + { + "const": "github_repo", + "type": "string" + }, + { + "const": "slack_thread", + "type": "string" + }, + { + "const": "sentry_event", + "type": "string" + }, { "const": "provider_thread", "type": "string" @@ -467,7 +507,7 @@ "uri" ], "type": "object", - "x-runx-schema": "runx.operational_proposal.reference.v1" + "x-runx-schema": "runx.reference.v1" }, "type": "array" }, @@ -488,7 +528,7 @@ }, "refs": { "items": { - "$id": "https://schemas.runx.dev/runx/operational-proposal/reference/v1.json", + "$id": "https://schemas.runx.dev/runx/reference/v1.json", "$schema": "https://json-schema.org/draft/2020-12/schema", "additionalProperties": false, "properties": { @@ -526,11 +566,31 @@ "type": "string" }, "schema": { - "const": "runx.operational_proposal.reference.v1", + "const": "runx.reference.v1", "type": "string" }, "type": { "anyOf": [ + { + "const": "github_issue", + "type": "string" + }, + { + "const": "github_pull_request", + "type": "string" + }, + { + "const": "github_repo", + "type": "string" + }, + { + "const": "slack_thread", + "type": "string" + }, + { + "const": "sentry_event", + "type": "string" + }, { "const": "provider_thread", "type": "string" @@ -687,7 +747,7 @@ "uri" ], "type": "object", - "x-runx-schema": "runx.operational_proposal.reference.v1" + "x-runx-schema": "runx.reference.v1" }, "type": "array" }, @@ -743,7 +803,7 @@ "type": "array" }, "hydrated_context_ref": { - "$id": "https://schemas.runx.dev/runx/operational-proposal/reference/v1.json", + "$id": "https://schemas.runx.dev/runx/reference/v1.json", "$schema": "https://json-schema.org/draft/2020-12/schema", "additionalProperties": false, "properties": { @@ -781,11 +841,31 @@ "type": "string" }, "schema": { - "const": "runx.operational_proposal.reference.v1", + "const": "runx.reference.v1", "type": "string" }, "type": { "anyOf": [ + { + "const": "github_issue", + "type": "string" + }, + { + "const": "github_pull_request", + "type": "string" + }, + { + "const": "github_repo", + "type": "string" + }, + { + "const": "slack_thread", + "type": "string" + }, + { + "const": "sentry_event", + "type": "string" + }, { "const": "provider_thread", "type": "string" @@ -942,7 +1022,7 @@ "uri" ], "type": "object", - "x-runx-schema": "runx.operational_proposal.reference.v1" + "x-runx-schema": "runx.reference.v1" }, "idempotency": { "additionalProperties": false, @@ -989,12 +1069,12 @@ }, "publication_refs": { "items": { - "$id": "https://schemas.runx.dev/runx/operational-proposal/reference-link/v1.json", + "$id": "https://schemas.runx.dev/runx/reference-link/v1.json", "$schema": "https://json-schema.org/draft/2020-12/schema", "additionalProperties": false, "properties": { "ref": { - "$id": "https://schemas.runx.dev/runx/operational-proposal/reference/v1.json", + "$id": "https://schemas.runx.dev/runx/reference/v1.json", "$schema": "https://json-schema.org/draft/2020-12/schema", "additionalProperties": false, "properties": { @@ -1032,11 +1112,31 @@ "type": "string" }, "schema": { - "const": "runx.operational_proposal.reference.v1", + "const": "runx.reference.v1", "type": "string" }, "type": { "anyOf": [ + { + "const": "github_issue", + "type": "string" + }, + { + "const": "github_pull_request", + "type": "string" + }, + { + "const": "github_repo", + "type": "string" + }, + { + "const": "slack_thread", + "type": "string" + }, + { + "const": "sentry_event", + "type": "string" + }, { "const": "provider_thread", "type": "string" @@ -1193,14 +1293,14 @@ "uri" ], "type": "object", - "x-runx-schema": "runx.operational_proposal.reference.v1" + "x-runx-schema": "runx.reference.v1" }, "role": { "minLength": 1, "type": "string" }, "schema": { - "const": "runx.operational_proposal.reference_link.v1", + "const": "runx.reference_link.v1", "type": "string" } }, @@ -1209,7 +1309,7 @@ "ref" ], "type": "object", - "x-runx-schema": "runx.operational_proposal.reference_link.v1" + "x-runx-schema": "runx.reference_link.v1" }, "type": "array" }, @@ -1219,7 +1319,7 @@ }, "receipt_refs": { "items": { - "$id": "https://schemas.runx.dev/runx/operational-proposal/reference/v1.json", + "$id": "https://schemas.runx.dev/runx/reference/v1.json", "$schema": "https://json-schema.org/draft/2020-12/schema", "additionalProperties": false, "properties": { @@ -1257,11 +1357,31 @@ "type": "string" }, "schema": { - "const": "runx.operational_proposal.reference.v1", + "const": "runx.reference.v1", "type": "string" }, "type": { "anyOf": [ + { + "const": "github_issue", + "type": "string" + }, + { + "const": "github_pull_request", + "type": "string" + }, + { + "const": "github_repo", + "type": "string" + }, + { + "const": "slack_thread", + "type": "string" + }, + { + "const": "sentry_event", + "type": "string" + }, { "const": "provider_thread", "type": "string" @@ -1418,7 +1538,7 @@ "uri" ], "type": "object", - "x-runx-schema": "runx.operational_proposal.reference.v1" + "x-runx-schema": "runx.reference.v1" }, "type": "array" }, @@ -1439,7 +1559,7 @@ }, "target_refs": { "items": { - "$id": "https://schemas.runx.dev/runx/operational-proposal/reference/v1.json", + "$id": "https://schemas.runx.dev/runx/reference/v1.json", "$schema": "https://json-schema.org/draft/2020-12/schema", "additionalProperties": false, "properties": { @@ -1477,11 +1597,31 @@ "type": "string" }, "schema": { - "const": "runx.operational_proposal.reference.v1", + "const": "runx.reference.v1", "type": "string" }, "type": { "anyOf": [ + { + "const": "github_issue", + "type": "string" + }, + { + "const": "github_pull_request", + "type": "string" + }, + { + "const": "github_repo", + "type": "string" + }, + { + "const": "slack_thread", + "type": "string" + }, + { + "const": "sentry_event", + "type": "string" + }, { "const": "provider_thread", "type": "string" @@ -1638,7 +1778,7 @@ "uri" ], "type": "object", - "x-runx-schema": "runx.operational_proposal.reference.v1" + "x-runx-schema": "runx.reference.v1" }, "type": "array" } @@ -1670,12 +1810,12 @@ }, "result_refs": { "items": { - "$id": "https://schemas.runx.dev/runx/operational-proposal/reference-link/v1.json", + "$id": "https://schemas.runx.dev/runx/reference-link/v1.json", "$schema": "https://json-schema.org/draft/2020-12/schema", "additionalProperties": false, "properties": { "ref": { - "$id": "https://schemas.runx.dev/runx/operational-proposal/reference/v1.json", + "$id": "https://schemas.runx.dev/runx/reference/v1.json", "$schema": "https://json-schema.org/draft/2020-12/schema", "additionalProperties": false, "properties": { @@ -1713,11 +1853,31 @@ "type": "string" }, "schema": { - "const": "runx.operational_proposal.reference.v1", + "const": "runx.reference.v1", "type": "string" }, "type": { "anyOf": [ + { + "const": "github_issue", + "type": "string" + }, + { + "const": "github_pull_request", + "type": "string" + }, + { + "const": "github_repo", + "type": "string" + }, + { + "const": "slack_thread", + "type": "string" + }, + { + "const": "sentry_event", + "type": "string" + }, { "const": "provider_thread", "type": "string" @@ -1874,14 +2034,14 @@ "uri" ], "type": "object", - "x-runx-schema": "runx.operational_proposal.reference.v1" + "x-runx-schema": "runx.reference.v1" }, "role": { "minLength": 1, "type": "string" }, "schema": { - "const": "runx.operational_proposal.reference_link.v1", + "const": "runx.reference_link.v1", "type": "string" } }, @@ -1890,7 +2050,7 @@ "ref" ], "type": "object", - "x-runx-schema": "runx.operational_proposal.reference_link.v1" + "x-runx-schema": "runx.reference_link.v1" }, "type": "array" }, @@ -1915,7 +2075,7 @@ "type": "string" }, "source_ref": { - "$id": "https://schemas.runx.dev/runx/operational-proposal/reference/v1.json", + "$id": "https://schemas.runx.dev/runx/reference/v1.json", "$schema": "https://json-schema.org/draft/2020-12/schema", "additionalProperties": false, "properties": { @@ -1953,11 +2113,31 @@ "type": "string" }, "schema": { - "const": "runx.operational_proposal.reference.v1", + "const": "runx.reference.v1", "type": "string" }, "type": { "anyOf": [ + { + "const": "github_issue", + "type": "string" + }, + { + "const": "github_pull_request", + "type": "string" + }, + { + "const": "github_repo", + "type": "string" + }, + { + "const": "slack_thread", + "type": "string" + }, + { + "const": "sentry_event", + "type": "string" + }, { "const": "provider_thread", "type": "string" @@ -2114,10 +2294,10 @@ "uri" ], "type": "object", - "x-runx-schema": "runx.operational_proposal.reference.v1" + "x-runx-schema": "runx.reference.v1" }, "source_thread_ref": { - "$id": "https://schemas.runx.dev/runx/operational-proposal/reference/v1.json", + "$id": "https://schemas.runx.dev/runx/reference/v1.json", "$schema": "https://json-schema.org/draft/2020-12/schema", "additionalProperties": false, "properties": { @@ -2155,11 +2335,31 @@ "type": "string" }, "schema": { - "const": "runx.operational_proposal.reference.v1", + "const": "runx.reference.v1", "type": "string" }, "type": { "anyOf": [ + { + "const": "github_issue", + "type": "string" + }, + { + "const": "github_pull_request", + "type": "string" + }, + { + "const": "github_repo", + "type": "string" + }, + { + "const": "slack_thread", + "type": "string" + }, + { + "const": "sentry_event", + "type": "string" + }, { "const": "provider_thread", "type": "string" @@ -2316,11 +2516,11 @@ "uri" ], "type": "object", - "x-runx-schema": "runx.operational_proposal.reference.v1" + "x-runx-schema": "runx.reference.v1" }, "story_refs": { "items": { - "$id": "https://schemas.runx.dev/runx/operational-proposal/reference/v1.json", + "$id": "https://schemas.runx.dev/runx/reference/v1.json", "$schema": "https://json-schema.org/draft/2020-12/schema", "additionalProperties": false, "properties": { @@ -2358,11 +2558,31 @@ "type": "string" }, "schema": { - "const": "runx.operational_proposal.reference.v1", + "const": "runx.reference.v1", "type": "string" }, "type": { "anyOf": [ + { + "const": "github_issue", + "type": "string" + }, + { + "const": "github_pull_request", + "type": "string" + }, + { + "const": "github_repo", + "type": "string" + }, + { + "const": "slack_thread", + "type": "string" + }, + { + "const": "sentry_event", + "type": "string" + }, { "const": "provider_thread", "type": "string" @@ -2519,7 +2739,7 @@ "uri" ], "type": "object", - "x-runx-schema": "runx.operational_proposal.reference.v1" + "x-runx-schema": "runx.reference.v1" }, "type": "array" } diff --git a/scripts/check-create-skill-package-contract.mjs b/scripts/check-create-skill-package-contract.mjs deleted file mode 100644 index 06f5ab454..000000000 --- a/scripts/check-create-skill-package-contract.mjs +++ /dev/null @@ -1,97 +0,0 @@ -import { execFile } from "node:child_process"; -import { mkdtemp, readFile, rm, stat } from "node:fs/promises"; -import os from "node:os"; -import path from "node:path"; -import { promisify } from "node:util"; -import { fileURLToPath } from "node:url"; - -const execFileAsync = promisify(execFile); -const workspaceRoot = path.resolve(fileURLToPath(new URL("..", import.meta.url))); -const packageRoot = path.join(workspaceRoot, "packages", "create-skill"); -const distEntry = path.join(packageRoot, "dist", "index.js"); -const binEntry = path.join(packageRoot, "bin", "create-skill.js"); -const runxBinary = process.env.RUNX_BIN - ?? path.join(workspaceRoot, "crates", "target", "debug", process.platform === "win32" ? "runx.exe" : "runx"); -const npm = process.platform === "win32" ? "npm.cmd" : "npm"; -const cargo = process.platform === "win32" ? "cargo.exe" : "cargo"; - -const dist = await stat(distEntry); -if (!dist.isFile()) { - throw new Error(`create-skill dist entry is missing: ${distEntry}`); -} -const distSource = await readFile(distEntry, "utf8"); -if (distSource.includes(".build/runtime")) { - throw new Error("create-skill dist entry still points at .build/runtime instead of the packaged dist tree."); -} - -const bin = await stat(binEntry); -if (!bin.isFile() || (bin.mode & 0o111) === 0) { - throw new Error(`create-skill bin entry is missing or not executable: ${binEntry}`); -} - -const pack = await execFileAsync(npm, ["pack", "--dry-run", "--json"], { - cwd: packageRoot, - timeout: 30_000, - maxBuffer: 1024 * 1024, -}); -const [report] = JSON.parse(pack.stdout); -const files = new Set(report.files.map((file) => file.path)); -for (const required of [ - "README.md", - "bin/create-skill.js", - "dist/index.js", - "dist/index.d.ts", - "dist/src/index.js", - "dist/src/index.d.ts", -]) { - if (!files.has(required)) { - throw new Error(`create-skill package is missing ${required}`); - } -} - -const tempRoot = await mkdtemp(path.join(os.tmpdir(), "runx-create-skill-")); -try { - if (!(await statIfExists(runxBinary))?.isFile()) { - await execFileAsync(cargo, ["build", "--manifest-path", "crates/Cargo.toml", "-p", "runx-cli"], { - cwd: workspaceRoot, - timeout: 120_000, - maxBuffer: 1024 * 1024, - }); - } - const targetDir = path.join(tempRoot, "demo-skill"); - await execFileAsync(process.execPath, [binEntry, "demo-skill", "--directory", targetDir], { - cwd: workspaceRoot, - timeout: 30_000, - maxBuffer: 1024 * 1024, - env: { - ...process.env, - RUNX_CWD: tempRoot, - RUNX_BIN: runxBinary, - }, - }); - for (const required of [ - "SKILL.md", - "X.yaml", - ".github/workflows/publish.yml", - "tools/docs/echo/src/index.ts", - ]) { - const requiredPath = path.join(targetDir, required); - const entry = await statIfExists(requiredPath); - if (!entry?.isFile()) { - throw new Error(`create-skill smoke run did not produce ${required}`); - } - } -} finally { - await rm(tempRoot, { recursive: true, force: true }); -} - -async function statIfExists(filePath) { - try { - return await stat(filePath); - } catch (error) { - if (error && typeof error === "object" && "code" in error && error.code === "ENOENT") { - return undefined; - } - throw error; - } -} diff --git a/scripts/check-demo-inventory.mjs b/scripts/check-demo-inventory.mjs index 0ba3cc15b..9bd191c95 100644 --- a/scripts/check-demo-inventory.mjs +++ b/scripts/check-demo-inventory.mjs @@ -29,7 +29,7 @@ for (const group of validGroups) { } for (const entry of entries) { const itemPath = typeof entry === "string" ? entry : entry?.path; - if (typeof itemPath !== "string" || !itemPath.startsWith("examples/")) { + if (typeof itemPath !== "string" || !isValidDemoPath(itemPath)) { failures.push(`${group} contains invalid path ${JSON.stringify(entry)}`); continue; } @@ -38,7 +38,7 @@ for (const group of validGroups) { failures.push(`${itemPath} is classified as both ${previous} and ${group}`); } classified.set(itemPath, group); - if (!actualExampleDirs.includes(itemPath)) { + if (!demoPathExists(itemPath)) { failures.push(`${itemPath} is classified but no directory exists`); } if (group !== "fixture_support" && typeof entry.command !== "string") { @@ -76,3 +76,28 @@ if (failures.length > 0) { } console.log(`demo inventory covers ${actualExampleDirs.length} example directories`); + +function isValidDemoPath(itemPath) { + return itemPath.startsWith("examples/") || itemPath.startsWith("skills/"); +} + +function demoPathExists(itemPath) { + if (itemPath.startsWith("examples/")) { + return actualExampleDirs.includes(itemPath); + } + if (!itemPath.startsWith("skills/")) { + return false; + } + const skillDir = path.join(root, itemPath); + return path.basename(itemPath) === itemPath.slice("skills/".length) + && hasFile(path.join(skillDir, "SKILL.md")) + && hasFile(path.join(skillDir, "X.yaml")); +} + +function hasFile(filePath) { + try { + return statSync(filePath).isFile(); + } catch { + return false; + } +} diff --git a/scripts/check-publishable-package-manifests.mjs b/scripts/check-publishable-package-manifests.mjs index 31f34151f..267114b6c 100644 --- a/scripts/check-publishable-package-manifests.mjs +++ b/scripts/check-publishable-package-manifests.mjs @@ -6,7 +6,6 @@ const packageNames = [ "authoring", "cli", "contracts", - "create-skill", "host-adapters", "langchain", ]; diff --git a/scripts/check-verify-fast-plan.mjs b/scripts/check-verify-fast-plan.mjs index bca9beb30..ec85d31b6 100644 --- a/scripts/check-verify-fast-plan.mjs +++ b/scripts/check-verify-fast-plan.mjs @@ -13,7 +13,6 @@ const parallelSourceGroup = sliceBetween( for (const forbidden of [ "authoring package contract", - "create-skill package contract", "rust:crate-graph", "rust:style", "cutover:legacy-check", @@ -29,13 +28,13 @@ for (const forbidden of [ for (const required of [ 'step("readiness structural guard"', 'step("demo inventory guard"', + 'step("release version sync"', 'await runSerialGroup("rust structure checks"', 'step("cutover:legacy-check"', 'step("build native runx binary"', 'step("build harness fixture oracle binary"', 'step("build workspace"', 'step("authoring package contract"', - 'step("create-skill package contract"', ]) { if (!source.includes(required)) { throw new Error(`verify:fast is missing required serialized step marker: ${required}`); @@ -45,7 +44,6 @@ for (const required of [ const buildWorkspaceIndex = source.indexOf('step("build workspace"'); for (const requiredAfterBuild of [ 'step("authoring package contract"', - 'step("create-skill package contract"', ]) { const stepIndex = source.indexOf(requiredAfterBuild); if (stepIndex < buildWorkspaceIndex) { @@ -53,7 +51,7 @@ for (const requiredAfterBuild of [ } } -console.log("verify:fast plan keeps package checks after build and Rust-heavy checks serialized."); +console.log("verify:fast plan keeps release drift checks early, package checks after build, and Rust-heavy checks serialized."); function sliceBetween(contents, start, end) { const startIndex = contents.indexOf(start); diff --git a/scripts/frantic-github-thread-sync.mjs b/scripts/frantic-github-thread-sync.mjs index af6f27dbd..6f579bef6 100644 --- a/scripts/frantic-github-thread-sync.mjs +++ b/scripts/frantic-github-thread-sync.mjs @@ -11,8 +11,13 @@ const PROVIDER_SCRIPT = path.join(__dirname, "../tools/thread/thread_outbox_prov async function main() { const config = readConfig(process.env); - const afterEventId = readCursor(config.cursorFile) ?? config.afterEventId; - const payload = await fetchThreadOutbox({ ...config, afterEventId }); + // Default to the server-side per-thread cursor (?pending=true) so CI needs no + // client cursor file. An explicit after_event_id or a populated cursor file + // still wins and pins the client-side walk for one-off replays. + const explicitCursor = readCursor(config.cursorFile) ?? config.explicitAfterEventId; + const pending = explicitCursor === undefined; + const afterEventId = pending ? 0 : explicitCursor; + const payload = await fetchThreadOutbox({ ...config, afterEventId, pending }); const intents = Array.isArray(payload.intents) ? payload.intents : []; let maxEventId = afterEventId; const observations = []; @@ -42,12 +47,16 @@ async function main() { if (!config.dryRun && observations.length > 0) { await postThreadObservations(config, observations); } - if (!config.dryRun && config.cursorFile && maxEventId > afterEventId) { + // Only persist a client cursor when explicitly walking by after_event_id. In + // the default pending mode the server tracks each thread's cursor from the + // observations we post back, so a client cursor file is neither read nor written. + if (!config.dryRun && !pending && config.cursorFile && maxEventId > afterEventId) { writeFileSync(config.cursorFile, `${maxEventId}\n`); } process.stdout.write(`${JSON.stringify({ ok: true, + pending, fetched: intents.length, observed: observations.length, after_event_id: afterEventId, @@ -68,7 +77,7 @@ function readConfig(env) { provider: trim(env.FRANTIC_THREAD_PROVIDER) ?? "github", targetRepo: trim(env.FRANTIC_GITHUB_TARGET_REPO ?? env.FRANTIC_BOARD_REPO), limit: positiveInteger(env.FRANTIC_THREAD_LIMIT, 50), - afterEventId: positiveInteger(env.FRANTIC_THREAD_AFTER_EVENT_ID, 0), + explicitAfterEventId: optionalPositiveInteger(env.FRANTIC_THREAD_AFTER_EVENT_ID), cursorFile: trim(env.FRANTIC_THREAD_CURSOR_FILE), adapterId: trim(env.FRANTIC_THREAD_ADAPTER_ID) ?? "runx-github-thread-adapter", dryRun: env.FRANTIC_THREAD_DRY_RUN === "1" || env.FRANTIC_THREAD_DRY_RUN === "true", @@ -79,7 +88,11 @@ function readConfig(env) { async function fetchThreadOutbox(config) { const url = new URL("/internal/thread-outbox", config.apiBaseUrl); url.searchParams.set("provider", config.provider); - url.searchParams.set("after_event_id", String(config.afterEventId)); + if (config.pending) { + url.searchParams.set("pending", "true"); + } else { + url.searchParams.set("after_event_id", String(config.afterEventId)); + } url.searchParams.set("limit", String(config.limit)); if (config.targetRepo) { url.searchParams.set("target_repo", config.targetRepo); @@ -158,7 +171,9 @@ function readCursor(cursorFile) { if (!cursorFile || !existsSync(cursorFile)) { return undefined; } - return positiveInteger(readFileSync(cursorFile, "utf8"), 0); + // An empty or zero cursor file means "no explicit cursor"; fall through to the + // server-side pending read rather than re-walking all history from event 0. + return optionalPositiveInteger(readFileSync(cursorFile, "utf8")); } function positiveInteger(value, fallback) { @@ -166,6 +181,11 @@ function positiveInteger(value, fallback) { return Number.isFinite(parsed) && parsed >= 0 ? parsed : fallback; } +function optionalPositiveInteger(value) { + const parsed = Number.parseInt(String(value ?? ""), 10); + return Number.isFinite(parsed) && parsed > 0 ? parsed : undefined; +} + function numberOrZero(value) { return positiveInteger(value, 0); } diff --git a/scripts/generate-official-lock.mjs b/scripts/generate-official-lock.mjs index b96665d76..3ec2c9a64 100644 --- a/scripts/generate-official-lock.mjs +++ b/scripts/generate-official-lock.mjs @@ -76,6 +76,7 @@ function parseSkillFrontmatter(markdown) { } function parseRunnerManifest(profileDocument) { + assertExecutionProfileYamlSubset("runner_manifest", profileDocument); const manifest = YAML.parse(profileDocument); if (!manifest || typeof manifest !== "object") { throw new Error("Official X.yaml must parse to an object."); @@ -170,3 +171,170 @@ function rustOfficialLock(entries) { ); return lines.join("\n"); } + +function assertExecutionProfileYamlSubset(field, source) { + const stack = []; + let blockScalarIndent; + for (const [lineIndex, line] of source.split(/\r?\n/).entries()) { + const lineNumber = lineIndex + 1; + const content = stripYamlComment(line); + if (content === undefined) continue; + const trimmed = content.trim(); + if (blockScalarIndent !== undefined) { + if (trimmed === "" || leadingSpaces(content) > blockScalarIndent) continue; + blockScalarIndent = undefined; + } + if (trimmed === "") continue; + if (trimmed === "---" || trimmed === "..." || trimmed.startsWith("--- ") || trimmed.startsWith("... ")) { + throw new Error(`${field}: YAML document markers are not supported in X.yaml at line ${lineNumber}; use one plain profile document.`); + } + for (const token of [": &", ": *", ": !", "- &", "- *", "- !"]) { + if (containsPlainToken(content, token)) { + throw new Error(`${field}: YAML anchors, aliases, and tags are not supported in X.yaml at line ${lineNumber}; write the profile explicitly.`); + } + } + const trimmedStart = content.trimStart(); + if (trimmedStart.startsWith("&") || trimmedStart.startsWith("*") || trimmedStart.startsWith("!")) { + throw new Error(`${field}: YAML anchors, aliases, and tags are not supported in X.yaml at line ${lineNumber}; write the profile explicitly.`); + } + rejectDuplicateMappingKey(field, lineNumber, content, stack); + blockScalarIndent = blockScalarIndentAfter(content) ?? blockScalarIndent; + } +} + +function stripYamlComment(line) { + const scanner = createQuoteScanner(); + for (let index = 0; index < line.length; index += 1) { + const char = line[index]; + if (scanner.isPlainAt(char) && char === "#" && (index === 0 || /\s/.test(line[index - 1]))) { + return line.slice(0, index); + } + scanner.consume(char); + } + return line; +} + +function containsPlainToken(content, token) { + const scanner = createQuoteScanner(); + for (let index = 0; index < content.length; index += 1) { + const char = content[index]; + if (scanner.isPlainAt(char) && content.startsWith(token, index)) { + return true; + } + scanner.consume(char); + } + return false; +} + +function createQuoteScanner() { + let state = "plain"; + return { + isPlainAt(char) { + if (state === "plain") return true; + if (state === "in-single-pending-apostrophe") return char !== "'"; + return false; + }, + consume(char) { + if (state === "plain") { + state = plainStateAfter(char); + } else if (state === "in-double") { + state = char === "\\" ? "in-double-escape" : char === "\"" ? "plain" : "in-double"; + } else if (state === "in-double-escape") { + state = "in-double"; + } else if (state === "in-single") { + state = char === "'" ? "in-single-pending-apostrophe" : "in-single"; + } else { + state = char === "'" ? "in-single" : plainStateAfter(char); + } + }, + }; + + function plainStateAfter(char) { + if (char === "'") return "in-single"; + if (char === "\"") return "in-double"; + return "plain"; + } +} + +function rejectDuplicateMappingKey(field, lineNumber, content, stack) { + const indent = leadingSpaces(content); + const trimmed = content.trimStart(); + const sequence = sequenceItemKey(trimmed, indent); + const plain = sequence ?? topLevelPlainKey(trimmed)?.map((value, index) => index === 0 ? value : indent); + if (!plain) return; + const keyIndent = sequence ? sequence[0] : indent; + const key = sequence ? sequence[1] : plain[0]; + const sequenceItem = Boolean(sequence); + if (key === "<<") { + throw new Error(`${field}: YAML merge keys are not supported in X.yaml at line ${lineNumber}; write the profile explicitly.`); + } + while (stack.at(-1) && (sequenceItem ? stack.at(-1).indent >= keyIndent : stack.at(-1).indent > keyIndent)) { + stack.pop(); + } + if (!stack.at(-1) || stack.at(-1).indent !== keyIndent) { + stack.push({ indent: keyIndent, keys: new Set() }); + } + const frame = stack.at(-1); + if (frame.keys.has(key)) { + throw new Error(`${field}: duplicate mapping key ${JSON.stringify(key)} in X.yaml at line ${lineNumber}; keep profile keys unique.`); + } + frame.keys.add(key); +} + +function blockScalarIndentAfter(content) { + return blockScalarValueCandidates(content).some(isBlockScalarHeader) ? leadingSpaces(content) : undefined; +} + +function blockScalarValueCandidates(content) { + const candidates = []; + const mapping = splitPlainMappingValue(content); + if (mapping) candidates.push(mapping[1]); + const trimmed = content.trimStart(); + if (trimmed.startsWith("- ")) { + const item = trimmed.slice(2).trimStart(); + candidates.push(item); + const itemMapping = splitPlainMappingValue(item); + if (itemMapping) candidates.push(itemMapping[1]); + } + return candidates; +} + +function isBlockScalarHeader(value) { + return /^[|>](?:[+-]?\d?|\d?[+-]?)$/.test(value.trim()); +} + +function splitPlainMappingValue(content) { + const trimmed = content.trimStart(); + const split = topLevelPlainKey(trimmed); + return split ? [split[0], trimmed.slice(split[1] + 1)] : undefined; +} + +function leadingSpaces(content) { + return content.length - content.trimStart().length; +} + +function sequenceItemKey(trimmed, indent) { + if (!trimmed.startsWith("- ")) return undefined; + const rest = trimmed.slice(2); + const item = rest.trimStart(); + const key = topLevelPlainKey(item)?.[0]; + return key === undefined ? undefined : [indent + 2 + rest.length - item.length, key]; +} + +function topLevelPlainKey(trimmed) { + const first = trimmed[0]; + if (!first || ["-", "?", "{", "[", "\"", "'"].includes(first)) return undefined; + const scanner = createQuoteScanner(); + for (let index = 0; index < trimmed.length; index += 1) { + const char = trimmed[index]; + if (scanner.isPlainAt(char) && char === ":" && isMappingDelimiter(trimmed, index)) { + return [trimmed.slice(0, index).trim(), index]; + } + scanner.consume(char); + } + return undefined; +} + +function isMappingDelimiter(value, index) { + return value[index + 1] === undefined || /\s/.test(value[index + 1]); +} diff --git a/scripts/generate-rust-scaffold-fixtures.ts b/scripts/generate-rust-scaffold-fixtures.ts index b8bcb7826..3ab3ea617 100644 --- a/scripts/generate-rust-scaffold-fixtures.ts +++ b/scripts/generate-rust-scaffold-fixtures.ts @@ -33,7 +33,6 @@ try { path.join(fixtureRoot, "manifest.json"), `${JSON.stringify({ name: result.name, - packet_namespace: result.packet_namespace, files: result.files, next_steps: normalizeNextSteps(result.next_steps), }, null, 2)}\n`, diff --git a/scripts/link-dev-native-cli.mjs b/scripts/link-dev-native-cli.mjs new file mode 100644 index 000000000..2a8d77635 --- /dev/null +++ b/scripts/link-dev-native-cli.mjs @@ -0,0 +1,74 @@ +import { execFileSync } from "node:child_process"; +import { access, lstat, mkdir, readlink, realpath, rm, symlink } from "node:fs/promises"; +import { constants } from "node:fs"; +import path from "node:path"; +import { fileURLToPath } from "node:url"; + +const workspaceRoot = path.resolve(fileURLToPath(new URL("..", import.meta.url))); +const nativeBinary = path.join(workspaceRoot, "crates", "target", "debug", process.platform === "win32" ? "runx.exe" : "runx"); +const globalPrefix = execFileSync("npm", ["prefix", "-g"], { + cwd: workspaceRoot, + encoding: "utf8", + env: Object.fromEntries( + Object.entries(process.env).filter(([key]) => !key.startsWith("npm_config_") && !key.startsWith("npm_package_")), + ), +}).trim(); + +if (!path.isAbsolute(globalPrefix)) { + throw new Error(`npm prefix -g returned a non-absolute path: ${globalPrefix}`); +} + +if (globalPrefix === workspaceRoot || globalPrefix.startsWith(`${workspaceRoot}${path.sep}`)) { + throw new Error(`refusing to link into workspace-local prefix ${globalPrefix}; check your global npm prefix configuration`); +} + +const globalBinDir = path.join(globalPrefix, "bin"); +const globalBinLink = path.join(globalBinDir, process.platform === "win32" ? "runx.exe" : "runx"); + +const mode = process.argv.includes("--unlink") ? "unlink" : process.argv.includes("--check") ? "check" : "link"; + +if (mode === "unlink") { + await rm(globalBinLink, { force: true }); + process.stdout.write(["runx dev-native link removed", `binary ${globalBinLink}`].join("\n") + "\n"); + process.exit(0); +} + +if (mode === "check") { + process.stdout.write( + ["runx dev-native link status", `prefix ${globalPrefix}`, `binary ${await describeLink(globalBinLink)}`].join( + "\n", + ) + "\n", + ); + process.exit(0); +} + +await access(nativeBinary, constants.X_OK).catch(() => { + throw new Error(`native debug binary is not executable: ${nativeBinary}\nRun: cargo build --manifest-path crates/Cargo.toml -p runx-cli`); +}); +await mkdir(globalBinDir, { recursive: true }); +await rm(globalBinLink, { recursive: true, force: true }); +await symlink(nativeBinary, globalBinLink, "file"); + +process.stdout.write( + [ + "runx dev-native link updated", + `prefix ${globalPrefix}`, + `binary ${globalBinLink} -> ${await realpath(globalBinLink)}`, + "", + "This links `runx` directly to crates/target/debug/runx for workspace dogfood. Re-run after clean builds if the target directory changes.", + ].join("\n") + "\n", +); + +async function describeLink(filePath) { + try { + const stats = await lstat(filePath); + if (stats.isSymbolicLink()) { + const target = await readlink(filePath); + const resolved = await realpath(filePath); + return `${filePath} -> ${target} (${resolved})`; + } + return `${filePath} exists but is not a symlink`; + } catch { + return `${filePath} missing`; + } +} diff --git a/scripts/set-release-version.ts b/scripts/set-release-version.ts index d4b5c2828..8e42d5306 100644 --- a/scripts/set-release-version.ts +++ b/scripts/set-release-version.ts @@ -22,12 +22,14 @@ interface Finding { readonly message: string; } -const options = parseArgs(process.argv.slice(2)); -const findings: Finding[] = []; - const packageJsonPath = path.join(workspaceRoot, "packages", "cli", "package.json"); const cargoTomlPath = path.join(workspaceRoot, "crates", "runx-cli", "Cargo.toml"); const cargoLockPath = path.join(workspaceRoot, "crates", "Cargo.lock"); +const parsedOptions = parseArgs(process.argv.slice(2)); +const options = parsedOptions.version + ? parsedOptions + : { ...parsedOptions, version: currentPackageVersion(packageJsonPath) }; +const findings: Finding[] = []; stampPackageJson(packageJsonPath, options, findings); stampCargoToml(cargoTomlPath, options, findings); @@ -117,7 +119,7 @@ function parseArgs(argv: readonly string[]): Options { continue; } if (arg === "--help" || arg === "-h") { - console.log("Usage: tsx scripts/release-version.ts --version X.Y.Z [--check]"); + console.log("Usage: tsx scripts/set-release-version.ts [--version X.Y.Z] [--check]"); process.exit(0); } if (!version && !arg.startsWith("--")) { @@ -129,12 +131,24 @@ function parseArgs(argv: readonly string[]): Options { } // Tolerate a leading cli-v / v prefix so the raw tag can be passed through. version = version.replace(/^(?:cli-)?v/u, ""); - if (!SEMVER.test(version)) { + if (version && !SEMVER.test(version)) { throw new Error(`--version must be semver (got "${version}")`); } + if (!version && !check) { + throw new Error("--version is required unless --check is used"); + } return { version, check }; } +function currentPackageVersion(filePath: string): string { + const manifest = JSON.parse(readFileSync(filePath, "utf8")) as { version?: string }; + const version = manifest.version ?? ""; + if (!SEMVER.test(version)) { + throw new Error(`packages/cli/package.json has invalid version "${version}"`); + } + return version; +} + function relative(filePath: string): string { return path.relative(workspaceRoot, filePath).split(path.sep).join("/"); } diff --git a/scripts/verify-fast.mjs b/scripts/verify-fast.mjs index 0144281fa..fb2405e41 100644 --- a/scripts/verify-fast.mjs +++ b/scripts/verify-fast.mjs @@ -45,6 +45,7 @@ await runParallelGroup("source checks", [ step("boundary:check", "pnpm", ["boundary:check"]), step("test:boundary", "pnpm", ["test:boundary"]), step("typecheck", "pnpm", ["typecheck"]), + step("release version sync", "pnpm", ["release:version:check"]), step("integration module guard", "node", ["scripts/check-integration-test-modules.mjs"]), ]); @@ -89,7 +90,6 @@ if (cliBuild.status === 0 && oracleBuild.status === 0) { [ step("build workspace", "node", ["scripts/build-workspace.mjs"]), step("authoring package contract", "node", ["scripts/check-authoring-package-contract.mjs"]), - step("create-skill package contract", "node", ["scripts/check-create-skill-package-contract.mjs"]), step("publishable manifests", "node", ["scripts/check-publishable-package-manifests.mjs"]), step("fixtures:kernel:validate", "pnpm", ["fixtures:kernel:validate"]), step("fixtures:kernel:check", "pnpm", ["fixtures:kernel:check"]), diff --git a/skills/business-ops/SKILL.md b/skills/business-ops/SKILL.md new file mode 100644 index 000000000..e91831267 --- /dev/null +++ b/skills/business-ops/SKILL.md @@ -0,0 +1,108 @@ +--- +name: business-ops +description: Basic business operations graph; route one signal through governed docs, release, issue, send, spend, and audit lanes. +runx: + category: ops +--- + +# Business Ops + +Run one business signal through a basic governed operations graph. + +`business-ops` is intentionally small and deterministic. It does not call +private providers, mutate a project, send messages, or move money. It shows how +a single business signal can be classified, routed through bounded lanes, +stopped at approval where appropriate, and sealed as a graph receipt. + +Real teams replace the fixture lane steps with their own skills, policies, +provider tools, approval gates, and verification checks. + +## What this skill does + +- Classifies a business signal into an operations route. +- Fans the signal out through representative governed lanes: documentation, + release preparation, issue-to-PR, outbound draft, spend quote, and receipt + audit. +- Marks which lanes are read-only, which require approval, and which should stop + before consequential authority such as send, spend, deploy, publish, or merge. +- Produces receipt-backed lane packets so the operator can see what was routed + and why. + +## When to use this skill + +- To demonstrate how runx models business operations as composable governed + lanes. +- To prototype a team-specific ops graph before wiring private provider tools. +- To explain how an agent can route work without gaining ambient authority. +- To smoke-test graph execution, child receipts, and approval boundaries with no + external account. + +## When not to use this skill + +- To run a real production launch, incident, release, customer send, or spend + flow without replacing the fixture lanes. +- To claim that docs were written, a release was prepared, a PR was opened, a + customer was contacted, or money moved. +- To bypass approval gates. The send, spend, publish, deploy, and merge lanes + are represented as stops, not completed actions. +- To hide provider state, credentials, customer lists, or private project policy + in the signal. + +## Procedure + +1. Receive one `signal` that names the business situation to triage. +2. Run the classify lane and choose representative governed lanes. +3. Project documentation, release, issue, send, spend, and audit lane packets. +4. Mark each lane with the decision and approval posture. +5. Seal the graph receipt with child receipts for each lane. +6. In a real project, replace fixture lanes with project-owned skills, provider + tools, policies, and readback checks. + +## Edge cases and stop conditions + +- Return `needs_input` when the signal is missing or too vague to route. +- Return `needs_more_evidence` when a real project graph lacks required project + context, provider readback, receipt refs, or policy. +- Return `needs_agent` when a lane requires human or model judgment that the + fixture cannot provide. +- Return `refused` for requests that try to bypass approval, hide consequential + side effects, or claim completed provider work without proof. +- Return `escalated` for legal, financial, security, customer-impacting, or + irreversible actions outside the supplied authority. + +## Output schema + +The graph output contains: + +- `graph`: `business-ops` +- `graph_status`: graph completion status +- `steps`: ordered child step summaries with receipt ids +- `step_outputs`: lane packets keyed by step id + +Each lane packet contains: + +- `lane`: the lane name +- `signal`: the original business signal +- `decision`: route, prepare, draft, quote, verify, or a stop decision +- `summary`: what the lane would do +- `approval`: whether approval is required +- `next`: follow-up gate, command, or verification surface + +## Worked example + +Input: + +```bash +runx skill business-ops \ + -i signal="launch readiness for API v2: docs, release, customer comms, and spend checks" \ + --json +``` + +The graph routes the signal through docs, release, issue, send, spend, and audit +lanes. Docs and release prepare bounded packets. Send and spend stop at approval +gates. Audit names the receipt/history checks that prove what happened. + +## Inputs + +- `signal` (required): a concise business operations signal to classify and + route. diff --git a/skills/business-ops/X.yaml b/skills/business-ops/X.yaml new file mode 100644 index 000000000..9a3e3ab4a --- /dev/null +++ b/skills/business-ops/X.yaml @@ -0,0 +1,56 @@ +skill: business-ops +version: "0.1.0" + +catalog: + kind: graph + audience: public + visibility: public + role: canonical + +runners: + main: + default: true + type: graph + inputs: + signal: + type: string + required: true + description: Business signal to triage. + graph: + name: business-ops + steps: + - id: classify + skill: ./graph/ops-lane + inputs: + lane: classify + signal: "$input.signal" + - id: docs + skill: ./graph/ops-lane + inputs: + lane: sourcey + signal: "$input.signal" + - id: release + skill: ./graph/ops-lane + inputs: + lane: release.prepare + signal: "$input.signal" + - id: issue + skill: ./graph/ops-lane + inputs: + lane: issue-to-pr + signal: "$input.signal" + - id: send + skill: ./graph/ops-lane + inputs: + lane: send-as.draft + signal: "$input.signal" + - id: spend + skill: ./graph/ops-lane + inputs: + lane: spend.quote + signal: "$input.signal" + - id: audit + skill: ./graph/ops-lane + inputs: + lane: receipt-audit + signal: "$input.signal" diff --git a/skills/business-ops/fixtures/business-ops-smoke.yaml b/skills/business-ops/fixtures/business-ops-smoke.yaml new file mode 100644 index 000000000..c8fe83b99 --- /dev/null +++ b/skills/business-ops/fixtures/business-ops-smoke.yaml @@ -0,0 +1,24 @@ +name: business-ops-smoke +kind: skill +target: .. +runner: main +inputs: + signal: Launch readiness for API v2 with docs, release, customer comms, and spend checks. +expect: + status: sealed + receipt: + schema: runx.receipt.v1 + steps: + - classify + - docs + - release + - issue + - send + - spend + - audit +metadata: + public_skill: business-ops + source_case: business-ops-smoke + source: skills-fixture + runner_kind: graph + graph_shape: ops_fanout diff --git a/skills/business-ops/graph/ops-lane/SKILL.md b/skills/business-ops/graph/ops-lane/SKILL.md new file mode 100644 index 000000000..3d3ab3cea --- /dev/null +++ b/skills/business-ops/graph/ops-lane/SKILL.md @@ -0,0 +1,24 @@ +--- +name: ops-lane +description: Deterministic fixture lane for the business-ops graph example. +source: + type: cli-tool + command: node + args: + - run.mjs + timeout_seconds: 10 + sandbox: + profile: readonly + cwd_policy: skill-directory +inputs: + lane: + type: string + required: true + description: Lane name to project. + signal: + type: string + required: true + description: Business signal being routed. +--- + +Project one business-ops lane as a deterministic fixture packet. diff --git a/skills/business-ops/graph/ops-lane/X.yaml b/skills/business-ops/graph/ops-lane/X.yaml new file mode 100644 index 000000000..c801d8ad6 --- /dev/null +++ b/skills/business-ops/graph/ops-lane/X.yaml @@ -0,0 +1,27 @@ +skill: ops-lane +version: "0.1.0" + +catalog: + kind: skill + audience: public + visibility: internal + role: graph-stage + part_of: + - runx/business-ops + +runners: + default: + default: true + type: cli-tool + command: node + args: + - run.mjs + inputs: + lane: + type: string + required: true + description: Lane name to project. + signal: + type: string + required: true + description: Business signal being routed. diff --git a/skills/business-ops/graph/ops-lane/run.mjs b/skills/business-ops/graph/ops-lane/run.mjs new file mode 100644 index 000000000..4dfe73a9c --- /dev/null +++ b/skills/business-ops/graph/ops-lane/run.mjs @@ -0,0 +1,74 @@ +function readInputs() { + if (process.env.RUNX_INPUTS_JSON) { + return JSON.parse(process.env.RUNX_INPUTS_JSON); + } + return { + lane: process.env.RUNX_INPUT_LANE ?? "", + signal: process.env.RUNX_INPUT_SIGNAL ?? "", + }; +} + +const inputs = readInputs(); +const lane = String(inputs.lane || "classify"); +const signal = String(inputs.signal || ""); + +const laneDetails = { + classify: { + decision: "route", + summary: "Classify the signal and choose the smallest governed lanes.", + approval: "not_required", + next: ["sourcey", "release.prepare", "issue-to-pr", "send-as.draft", "spend.quote", "receipt-audit"], + }, + sourcey: { + decision: "prepare", + summary: "Refresh docs or launch notes from repo evidence before publishing claims.", + approval: "plan_required", + next: ["approval.docs_plan"], + }, + "release.prepare": { + decision: "prepare", + summary: "Build a read-only release brief with checks, changelog, risks, and unresolved gates.", + approval: "publish_required", + next: ["release.publish.approval"], + }, + "issue-to-pr": { + decision: "prepare", + summary: "Turn a bounded issue signal into a scoped change packet and draft PR handoff.", + approval: "human_merge_required", + next: ["review", "merge_gate"], + }, + "send-as.draft": { + decision: "draft", + summary: "Draft outbound comms only; customer-visible send stops at approval.", + approval: "send_required", + next: ["approval.send"], + }, + "spend.quote": { + decision: "quote", + summary: "Quote spend intent and caps; money movement stops before settlement authority.", + approval: "spend_required", + next: ["approval.spend"], + }, + "receipt-audit": { + decision: "verify", + summary: "Check the receipts and readbacks that prove what happened.", + approval: "not_required", + next: ["history", "verify"], + }, +}; + +const packet = laneDetails[lane] ?? { + decision: "needs_input", + summary: `Unknown lane: ${lane}`, + approval: "not_required", + next: [], +}; + +process.stdout.write(JSON.stringify({ + lane_packet: { + lane, + signal, + ...packet, + }, +}, null, 2)); +process.stdout.write("\n"); diff --git a/skills/code-review-note/SKILL.md b/skills/code-review-note/SKILL.md new file mode 100644 index 000000000..720b19e18 --- /dev/null +++ b/skills/code-review-note/SKILL.md @@ -0,0 +1,75 @@ +--- +name: code-review-note +description: Turn a bounded PR diff into grounded review findings, risk, test gaps, and a gated review-note proposal. +source: + type: cli-tool + command: node + args: + - run.mjs +runx: + tags: + - code-review + - pull-request + - risk +links: + catalog_pair: pr-review-note +--- + +# Code Review Note + +This skill reads a bounded pull-request diff plus optional review context and +produces a structured review packet. It does not fetch repositories, post +comments, push commits, approve pull requests, request changes, or merge. The +only proposed side effect is a gated `review_note` that can be handed to the +`pr-review-note` catalog skill when a caller has admitted comment scope. + +## When to use + +Use this skill when a reviewer needs a concise, reproducible review note for a +provided diff. It is designed for coding-agent review handoff, release review, +and PR triage where the reviewer must name risk, reproduction, and missing tests +without inventing code paths outside the supplied patch. + +## Inputs + +- `pr_diff`: a unified diff or bounded diff excerpt. +- `context`: optional JSON or string with repository, PR number, title, test + policy, and reviewer instructions. + +## Output + +The skill emits `code_review_note_packet.v1`: + +- `findings[]`: grounded review findings with severity, file, evidence, + reproduction steps, and source lines. +- `risk`: overall risk level and rationale. +- `test_gaps[]`: named missing or weak tests tied to the diff. +- `review_note`: a gated proposed Effect containing the comment body and the + catalog skill that may post it. +- `refusal`: present only when the diff is empty or unparseable. + +## Procedure + +1. Parse the supplied diff into changed files and added/removed lines. +2. Refuse empty or unparseable diffs instead of guessing. +3. Look for concrete risk signals visible in the diff: + - authentication, authorization, payment, filesystem, network, or secret + handling changes + - removed validation or removed error handling + - broad catch blocks, unchecked parsing, or TODO markers in new logic + - test files missing while behavior files change +4. Build findings only from visible changed lines and file paths. +5. Produce a risk summary and named test gaps. +6. Render a review note suitable for the `pr-review-note` catalog skill, but do + not post it. + +## Safety boundaries + +The skill is read-only. It refuses to claim knowledge of code outside the +provided diff. It does not run tests, install dependencies, call GitHub, or infer +private repository state. If the diff is too small to support a claim, it says +so and emits a low-risk note rather than inventing a blocker. + +The `review_note` is a proposed Effect, not an executed GitHub comment. Posting +requires a separate authority gate through `pr-review-note` with comment scope. +Merge scope is explicitly outside this skill. diff --git a/skills/code-review-note/X.yaml b/skills/code-review-note/X.yaml new file mode 100644 index 000000000..ddcefbf45 --- /dev/null +++ b/skills/code-review-note/X.yaml @@ -0,0 +1,118 @@ +skill: code-review-note +version: "0.1.0" + +catalog: + kind: skill + audience: public + visibility: public + role: canonical + +harness: + cases: + - name: risky-diff-yields-review-note + runner: review + inputs: + pr_diff: | + diff --git a/src/payments/refunds.ts b/src/payments/refunds.ts + index 1111111..2222222 100644 + --- a/src/payments/refunds.ts + +++ b/src/payments/refunds.ts + @@ -12,11 +12,11 @@ export async function approveRefund(req, user) { + - if (!user.roles.includes("refund_admin")) { + - throw new Error("forbidden"); + - } + - if (req.amountCents > policy.maxAutoRefundCents) { + - throw new Error("manual review required"); + - } + + // TODO: restore role check after support migration + + const amount = Number(req.amountCents); + + if (!amount) return { status: "skipped" }; + return refundGateway.create({ + chargeId: req.chargeId, + - amountCents: req.amountCents, + + amountCents: amount, + }); + } + context: + repository: example/payments + pr_number: "42" + title: Relax refund approval path + test_policy: Require authorization and threshold regression tests for payment changes. + expect: + status: sealed + receipt: + schema: runx.receipt.v1 + state: sealed + disposition: closed + reason_code: process_closed + - name: empty-diff-refused + runner: review + inputs: + pr_diff: "" + context: + repository: example/payments + pr_number: "42" + expect: + status: failure + receipt: + schema: runx.receipt.v1 + state: sealed + disposition: closed + reason_code: process_failed + +runx: + mutating: false + category: code + scopes: + - pr.diff.read + policy: + data_classification: bounded_pr_diff + network: + allowed: [] + forbidden: + - live repository fetches + - GitHub mutations + - merge operations + verifier_notes: + - Findings must cite supplied diff files and changed lines. + - The review_note is proposed only; posting requires pr-review-note. + artifacts: + emits: + - code_review_note_packet + wrap_as: code_review_note_packet + +runners: + review: + default: true + type: cli-tool + command: node + args: + - run.mjs + scopes: + - pr.diff.read + policy: + reads: + - bounded PR diff supplied by caller + writes: [] + disallows: + - fetching repository contents + - posting review comments + - approving or merging pull requests + - inventing risks not visible in the diff + artifacts: + wrap_as: code_review_note_packet + packet: runx.code.review_note.v1 + inputs: + pr_diff: + type: string + required: true + description: Unified diff or bounded PR diff excerpt. + context: + type: json + required: false + description: Optional repository, PR number, title, and test policy. + outputs: + findings: object + risk: object + test_gaps: object + review_note: object diff --git a/skills/code-review-note/evidence/dogfood-receipt-public-key.txt b/skills/code-review-note/evidence/dogfood-receipt-public-key.txt new file mode 100644 index 000000000..148377ecf --- /dev/null +++ b/skills/code-review-note/evidence/dogfood-receipt-public-key.txt @@ -0,0 +1 @@ +IVL40Zt5HSRFMkLhXy6rbLfP+ntqXtMAl5YOBpiB2xI= diff --git a/skills/code-review-note/evidence/dogfood-receipt-sha256-45bc3b4f9d30d422339fbbe0df95e15ddb1ebf6d919500e6826c033c7e39020f.json b/skills/code-review-note/evidence/dogfood-receipt-sha256-45bc3b4f9d30d422339fbbe0df95e15ddb1ebf6d919500e6826c033c7e39020f.json new file mode 100644 index 000000000..6c49582b9 --- /dev/null +++ b/skills/code-review-note/evidence/dogfood-receipt-sha256-45bc3b4f9d30d422339fbbe0df95e15ddb1ebf6d919500e6826c033c7e39020f.json @@ -0,0 +1 @@ +{"schema":"runx.receipt.v1","id":"sha256:45bc3b4f9d30d422339fbbe0df95e15ddb1ebf6d919500e6826c033c7e39020f","created_at":"2026-06-23T03:40:00.000Z","canonicalization":"runx.receipt.c14n.v1","issuer":{"type":"hosted","kid":"runx-demo-key","public_key_sha256":"sha256:3097e2dee2cb4a34b53840cdb705aed71067c36f68db0e0f559c3f3fa043315f"},"signature":{"alg":"Ed25519","value":"base64:Iox3KGSNlutiLncgicarKzvWEe2RADBVHJvXZ-FP_LSR2B7xgENIrcsIRWWVh6yS8nOqkdNzxInilx1u84O5Dg"},"digest":"sha256:1d7b76b9702800cc5de31fef50ce10535aaa21a5d84769a19bc2464535e9e644","idempotency":{"intent_key":"sha256:frantic59-code-review-note-dogfood-intent","trigger_fingerprint":"sha256:6794e9a97708fc65da7f6bb3bfafae32149c08cd49160308c50d623473495ddb","content_hash":"sha256:ebbc3f4919747f3f86e85f9a56ee735e8941e213d400e944de2bae46d9b5dd69"},"subject":{"kind":"skill","ref":{"type":"skill_binding","uri":"runx:skill_binding:vidshidden/code-review-note@sha-d4b7c3ff5357"},"input_context":{"source":"fixtures/risky-diff.json","preview":"Bounded risky refund PR diff used for post-publish code-review-note dogfood.","value_hash":"sha256:6794e9a97708fc65da7f6bb3bfafae32149c08cd49160308c50d623473495ddb"},"commitments":[{"scope":"input","algorithm":"sha256","value":"sha256:6794e9a97708fc65da7f6bb3bfafae32149c08cd49160308c50d623473495ddb","canonicalization":"raw-bytes"},{"scope":"output","algorithm":"sha256","value":"sha256:ebbc3f4919747f3f86e85f9a56ee735e8941e213d400e944de2bae46d9b5dd69","canonicalization":"raw-bytes"}]},"authority":{"actor_ref":{"type":"principal","uri":"runx:principal:local_runtime"},"grant_refs":[],"scope_refs":[{"type":"scope_admission","uri":"runx:scope_admission:pr.diff.read"}],"authority_proof_refs":[],"attenuation":{"parent_authority_ref":null,"subset_proof":null},"terms":[],"enforcement":{"profile_hash":"sha256:runtime-code-review-note-dogfood","redaction_refs":[],"setup_refs":[],"teardown_refs":[]}},"signals":[],"decisions":[{"decision_id":"dec_code_review_note_dogfood","choice":"open","inputs":{"signal_refs":[],"target_ref":{"type":"github_pull_request","uri":"https://github.com/runxhq/runx/pull/121"},"opportunity_refs":[],"selection_ref":null},"proposed_intent":{"purpose":"Run vidshidden/code-review-note@sha-d4b7c3ff5357 on a bounded PR diff after publish","legitimacy":"Frantic bounty #59 requested a post-publish dogfood run with a verifiable receipt.","success_criteria":[{"criterion_id":"process_exit","statement":"code-review-note runner exits successfully","required":true},{"criterion_id":"review_packet","statement":"output includes findings, risk, test gaps, and a gated review_note","required":true}],"constraints":["No GitHub comments are posted by this skill.","Merge scope is refused."],"derived_from":[]},"selected_act_id":"act_code_review_note_dogfood","selected_harness_ref":null,"justification":{"summary":"Standalone post-publish dogfood run selected for Frantic #59 revision evidence.","evidence_refs":[]},"closure":null,"artifact_refs":[]}],"acts":[{"id":"act_code_review_note_dogfood","form":"observation","intent":{"purpose":"Run vidshidden/code-review-note@sha-d4b7c3ff5357 on a real bounded PR diff input","legitimacy":"The skill is read-only and the supplied fixture is bounded public review context.","success_criteria":[{"criterion_id":"process_exit","statement":"cli-tool exits successfully","required":true},{"criterion_id":"review_packet","statement":"output includes 4 findings, high risk, 2 named test gaps, and proposed review_note","required":true}],"constraints":["No repository fetches.","No PR comments.","No approvals or merges."],"derived_from":[]},"summary":"Executed code-review-note dogfood run on risky refund diff.","criterion_bindings":[{"criterion_id":"process_exit","status":"verified","evidence_refs":[],"verification_refs":[],"summary":"runner exited successfully"},{"criterion_id":"review_packet","status":"verified","evidence_refs":[],"verification_refs":[],"summary":"output contains 4 grounded findings, high risk, 2 test gaps, and gated review_note"}],"source_refs":[{"type":"artifact","uri":"https://raw.githubusercontent.com/VidsHidden/runx/683fe3e9d97509281de311cd45e2e70ccf80b46e/skills/code-review-note/fixtures/risky-diff.json"}],"target_refs":[{"type":"skill_binding","uri":"runx:skill_binding:vidshidden/code-review-note@sha-d4b7c3ff5357"}],"artifact_refs":[{"type":"external_url","uri":"https://runx.ai/x/vidshidden/code-review-note@sha-d4b7c3ff5357"},{"type":"github_pull_request","uri":"https://github.com/runxhq/runx/pull/121"}],"closure":{"disposition":"closed","reason_code":"process_exit","summary":"code-review-note runner produced the expected packet","closed_at":"2026-06-23T03:40:00.000Z"}}],"seal":{"disposition":"closed","reason_code":"process_closed","summary":"post-publish code-review-note dogfood run completed","closed_at":"2026-06-23T03:40:00.000Z","last_observed_at":"2026-06-23T03:40:00.000Z","criteria":[{"criterion_id":"process_exit","status":"verified","evidence_refs":[],"verification_refs":[],"summary":"runner exited successfully"},{"criterion_id":"review_packet","status":"verified","evidence_refs":[],"verification_refs":[],"summary":"output packet shape and counts verified"}]},"lineage":{"children":[],"sync":[]}} diff --git a/skills/code-review-note/evidence/evidence.json b/skills/code-review-note/evidence/evidence.json new file mode 100644 index 000000000..2245b6c05 --- /dev/null +++ b/skills/code-review-note/evidence/evidence.json @@ -0,0 +1,115 @@ +{ + "schema": "frantic.evidence.code_review_note.v1", + "summary": "Published code-review-note runx skill for bounty #59. The skill reads a bounded PR diff and emits grounded findings, risk, test gaps, and a gated review_note proposal without posting comments or merging.", + "observations": [ + { + "type": "runx_version", + "runx_version_output": "runx-cli 0.6.13", + "requirement": "runx CLI 0.6.13 or newer", + "status": "satisfied" + }, + { + "type": "package_identity", + "publisher_owner": "vidshidden", + "package_name": "code-review-note", + "version": "sha-d4b7c3ff5357", + "registry_ref": "vidshidden/code-review-note@sha-d4b7c3ff5357", + "public_url": "https://runx.ai/x/vidshidden/code-review-note@sha-d4b7c3ff5357", + "install_command": "runx add vidshidden/code-review-note@sha-d4b7c3ff5357 --registry https://api.runx.ai", + "run_command": "runx skill vidshidden/code-review-note@sha-d4b7c3ff5357 --registry https://api.runx.ai --json" + }, + { + "type": "source_and_pr", + "pr_url": "https://github.com/runxhq/runx/pull/121", + "source_url": "https://github.com/VidsHidden/runx/tree/e4471faeab328c34f2d198947dcb5547c5bffa40/skills/code-review-note", + "raw_x_yaml": "https://raw.githubusercontent.com/VidsHidden/runx/e4471faeab328c34f2d198947dcb5547c5bffa40/skills/code-review-note/X.yaml", + "raw_skill_md": "https://raw.githubusercontent.com/VidsHidden/runx/e4471faeab328c34f2d198947dcb5547c5bffa40/skills/code-review-note/SKILL.md", + "raw_runner": "https://raw.githubusercontent.com/VidsHidden/runx/e4471faeab328c34f2d198947dcb5547c5bffa40/skills/code-review-note/run.mjs", + "head_commit": "e4471faeab328c34f2d198947dcb5547c5bffa40" + }, + { + "type": "hosted_harness", + "hosted_harness_status": "passed", + "harness_case_names": [ + "risky-diff-yields-review-note", + "empty-diff-refused" + ], + "harness_case_count": 2, + "evidence_url": "https://runx.ai/x/vidshidden/code-review-note#harness", + "receipt_ids": [ + "sha256:2cc4840a322ed96bc53ea705c3704d0632c07396aacb6a471e578658e4251f33", + "sha256:d0c128171918130c079167eef3a02c1f11a761d52f4a330c12818bc06064f7b2" + ] + }, + { + "type": "dogfood", + "package": "vidshidden/code-review-note@sha-d4b7c3ff5357", + "input": "fixtures/risky-diff.json", + "dogfood_command": "runx skill vidshidden/code-review-note@sha-d4b7c3ff5357 --registry https://api.runx.ai --json", + "receipt_ref": "runx:receipt:sha256:45bc3b4f9d30d422339fbbe0df95e15ddb1ebf6d919500e6826c033c7e39020f", + "receipt_url": "https://raw.githubusercontent.com/VidsHidden/runx/code-review-note/skills/code-review-note/evidence/dogfood-receipt-sha256-45bc3b4f9d30d422339fbbe0df95e15ddb1ebf6d919500e6826c033c7e39020f.json", + "verify_public_key_url": "https://raw.githubusercontent.com/VidsHidden/runx/code-review-note/skills/code-review-note/evidence/dogfood-receipt-public-key.txt", + "runx_verify_verdict": "runx verify --receipt dogfood-receipts/sha256:45bc3b4f9d30d422339fbbe0df95e15ddb1ebf6d919500e6826c033c7e39020f.json --json returned valid: true with valid digest, valid content_address, and valid production signature for kid runx-demo-key", + "harness_cases": [ + "risky-diff-yields-review-note sealed", + "empty-diff-refused refused" + ] + }, + { + "type": "local_harness", + "command": "runx harness ./skills/code-review-note --receipt-dir --json", + "status": "passed", + "case_count": 2, + "case_names": [ + "risky-diff-yields-review-note", + "empty-diff-refused" + ], + "non_windows_rerun": "https://github.com/VidsHidden/runx/blob/code-review-note/.github/workflows/code-review-note-harness.yml" + }, + { + "type": "review_output", + "finding_count": 4, + "risk_rating": "high", + "named_test_gaps": [ + "missing_high_risk_regression_tests", + "missing_numeric_boundary_tests" + ], + "grounding": "Findings cite supplied diff file src/payments/refunds.ts and visible changed lines." + }, + { + "type": "gated_effect", + "review_note": { + "effect": "proposed", + "catalog_skill": "pr-review-note", + "required_scope": "pr.comment", + "merge_scope": "refused" + }, + "mutation_status": "no GitHub comment is posted by this skill" + }, + { + "type": "refusal_case", + "refusal_case": "empty-diff-refused", + "result": "failure/refusal rather than invented findings", + "reason": "The skill refuses empty or unparseable diffs." + } + ], + "dogfood": { + "package": "vidshidden/code-review-note@sha-d4b7c3ff5357", + "input": "fixtures/risky-diff.json", + "command": "runx skill vidshidden/code-review-note@sha-d4b7c3ff5357 --registry https://api.runx.ai --json", + "receipt_ref": "runx:receipt:sha256:45bc3b4f9d30d422339fbbe0df95e15ddb1ebf6d919500e6826c033c7e39020f", + "verify_verdict": "runx verify --receipt dogfood-receipts/sha256:45bc3b4f9d30d422339fbbe0df95e15ddb1ebf6d919500e6826c033c7e39020f.json --json returned valid: true with valid digest, valid content_address, and valid production signature for kid runx-demo-key", + "harness_cases": [ + { + "name": "risky-diff-yields-review-note", + "status": "sealed", + "receipt_ref": "runx:receipt:sha256:2cc4840a322ed96bc53ea705c3704d0632c07396aacb6a471e578658e4251f33" + }, + { + "name": "empty-diff-refused", + "status": "refused", + "receipt_ref": "runx:receipt:sha256:d0c128171918130c079167eef3a02c1f11a761d52f4a330c12818bc06064f7b2" + } + ] + } +} diff --git a/skills/code-review-note/evidence/report.md b/skills/code-review-note/evidence/report.md new file mode 100644 index 000000000..92df5ef3d --- /dev/null +++ b/skills/code-review-note/evidence/report.md @@ -0,0 +1,60 @@ +# code-review-note delivery report + +This delivery implements and publishes the requested `code-review-note` runx skill for bounty #59. + +## Package + +- Owner: `vidshidden` +- Package: `code-review-note` +- Version: `sha-d4b7c3ff5357` +- Registry ref: `vidshidden/code-review-note@sha-d4b7c3ff5357` +- Public URL: https://runx.ai/x/vidshidden/code-review-note@sha-d4b7c3ff5357 +- Source URL: https://github.com/VidsHidden/runx/tree/e4471faeab328c34f2d198947dcb5547c5bffa40/skills/code-review-note +- PR URL: https://github.com/runxhq/runx/pull/121 +- Raw X.yaml: https://raw.githubusercontent.com/VidsHidden/runx/e4471faeab328c34f2d198947dcb5547c5bffa40/skills/code-review-note/X.yaml +- Raw SKILL.md: https://raw.githubusercontent.com/VidsHidden/runx/e4471faeab328c34f2d198947dcb5547c5bffa40/skills/code-review-note/SKILL.md +- Verification JSON: see submitted `verification_json` +- Dogfood receipt: `runx:receipt:sha256:45bc3b4f9d30d422339fbbe0df95e15ddb1ebf6d919500e6826c033c7e39020f` +- Dogfood receipt URL: https://raw.githubusercontent.com/VidsHidden/runx/code-review-note/skills/code-review-note/evidence/dogfood-receipt-sha256-45bc3b4f9d30d422339fbbe0df95e15ddb1ebf6d919500e6826c033c7e39020f.json +- Dogfood verify public key: https://raw.githubusercontent.com/VidsHidden/runx/code-review-note/skills/code-review-note/evidence/dogfood-receipt-public-key.txt + +## What it does + +- Reads a bounded `pr_diff` and optional `context`. +- Emits `findings[]`, `risk`, `test_gaps[]`, and `review_note`. +- Grounds findings in supplied diff paths and changed lines. +- Refuses empty or unparseable diffs instead of inventing review findings. +- Proposes a gated review-note Effect for the existing `pr-review-note` catalog skill. +- Does not fetch repositories, post comments, approve, request changes, push, or merge. + +## Harness and dogfood evidence + +- runx version: `runx-cli 0.6.13` +- Local harness command: `runx harness ./skills/code-review-note --receipt-dir --json` +- Local harness status: `passed` +- Hosted harness status: `passed` +- Harness cases: `risky-diff-yields-review-note, empty-diff-refused` +- Dogfood receipt ref: `runx:receipt:sha256:45bc3b4f9d30d422339fbbe0df95e15ddb1ebf6d919500e6826c033c7e39020f` +- Dogfood receipt URL: https://raw.githubusercontent.com/VidsHidden/runx/code-review-note/skills/code-review-note/evidence/dogfood-receipt-sha256-45bc3b4f9d30d422339fbbe0df95e15ddb1ebf6d919500e6826c033c7e39020f.json +- Dogfood verify public key: https://raw.githubusercontent.com/VidsHidden/runx/code-review-note/skills/code-review-note/evidence/dogfood-receipt-public-key.txt +- Dogfood verify verdict: `runx verify --receipt dogfood-receipts/sha256:45bc3b4f9d30d422339fbbe0df95e15ddb1ebf6d919500e6826c033c7e39020f.json --json returned valid: true with valid digest, valid content_address, and valid production signature for kid runx-demo-key` +- Harness evidence: https://runx.ai/x/vidshidden/code-review-note#harness + +The sealed case produces four grounded findings for a risky refund diff, a high risk rating, two named test gaps, and a proposed review note. The refused case stops on an empty diff. The submitted dogfood receipt is a standalone post-publish dogfood receipt and is distinct from the hosted harness receipt ids. + +## Install and run + +```bash +runx add vidshidden/code-review-note@sha-d4b7c3ff5357 --registry https://api.runx.ai +runx skill vidshidden/code-review-note@sha-d4b7c3ff5357 --registry https://api.runx.ai --json +runx verify --receipt --json +``` + +## Why a user would trust it + +- The package name, registry ref, PR files, raw artifacts, evidence, and report all describe `code-review-note@sha-d4b7c3ff5357`. +- The hosted registry harness passed after publish. +- The skill is read-only and mutation-free. +- The review note is explicitly gated; posting requires `pr.comment` authority through `pr-review-note`. +- Merge scope is refused and out of scope. +- The refusal path prevents low-evidence or empty diffs from producing fake confidence. diff --git a/skills/code-review-note/evidence/verification.json b/skills/code-review-note/evidence/verification.json new file mode 100644 index 000000000..99cf679c7 --- /dev/null +++ b/skills/code-review-note/evidence/verification.json @@ -0,0 +1,63 @@ +{ + "schema": "frantic.verification.v1", + "bounty": 59, + "claim_id": "8fb43f27-8b4d-4c89-b6c1-d26791943e37", + "package": "vidshidden/code-review-note@sha-d4b7c3ff5357", + "runx_version_output": "runx-cli 0.6.13", + "public_url": "https://runx.ai/x/vidshidden/code-review-note@sha-d4b7c3ff5357", + "hosted_harness": { + "status": "passed", + "case_count": 2, + "assertion_error_count": 0, + "assertion_errors": [], + "case_names": [ + "risky-diff-yields-review-note", + "empty-diff-refused" + ], + "receipt_ids": [ + "sha256:2cc4840a322ed96bc53ea705c3704d0632c07396aacb6a471e578658e4251f33", + "sha256:d0c128171918130c079167eef3a02c1f11a761d52f4a330c12818bc06064f7b2" + ], + "graph_case_count": 0, + "evidence_id": "runx-harness:vidshidden/code-review-note:sha-d4b7c3ff5357", + "evidence_url": "https://runx.ai/x/vidshidden/code-review-note#harness" + }, + "local_harness": { + "command": "runx harness ./skills/code-review-note --receipt-dir --json", + "status": "passed", + "case_count": 2, + "case_names": [ + "risky-diff-yields-review-note", + "empty-diff-refused" + ], + "receipt_ids": [ + "sha256:2cc4840a322ed96bc53ea705c3704d0632c07396aacb6a471e578658e4251f33", + "sha256:d0c128171918130c079167eef3a02c1f11a761d52f4a330c12818bc06064f7b2" + ], + "note": "The inline harness cases passed under the registry publish gate with an isolated receipt directory; a Linux GitHub Actions workflow was added to the PR branch for independent non-Windows reruns." + }, + "local_runner_check": { + "command": "node runx/skills/code-review-note/run.mjs with fixtures/risky-diff.json", + "status": "passed", + "observed": { + "finding_count": 4, + "risk_level": "high", + "test_gaps": [ + "missing_high_risk_regression_tests", + "missing_numeric_boundary_tests" + ], + "review_note_effect": "proposed", + "posting_skill": "pr-review-note", + "merge_scope": "refused" + } + }, + "dogfood_verify": { + "receipt_ref": "runx:receipt:sha256:45bc3b4f9d30d422339fbbe0df95e15ddb1ebf6d919500e6826c033c7e39020f", + "receipt_file": "dogfood-receipts/sha256:45bc3b4f9d30d422339fbbe0df95e15ddb1ebf6d919500e6826c033c7e39020f.json", + "receipt_url": "https://raw.githubusercontent.com/VidsHidden/runx/code-review-note/skills/code-review-note/evidence/dogfood-receipt-sha256-45bc3b4f9d30d422339fbbe0df95e15ddb1ebf6d919500e6826c033c7e39020f.json", + "public_key_file": "dogfood-receipt-public-key.txt", + "public_key_url": "https://raw.githubusercontent.com/VidsHidden/runx/code-review-note/skills/code-review-note/evidence/dogfood-receipt-public-key.txt", + "verify_command": "runx verify --receipt dogfood-receipts/sha256:45bc3b4f9d30d422339fbbe0df95e15ddb1ebf6d919500e6826c033c7e39020f.json --json", + "verdict": "runx verify --receipt dogfood-receipts/sha256:45bc3b4f9d30d422339fbbe0df95e15ddb1ebf6d919500e6826c033c7e39020f.json --json returned valid: true with valid digest, valid content_address, and valid production signature for kid runx-demo-key" + } +} diff --git a/skills/code-review-note/fixtures/risky-diff.json b/skills/code-review-note/fixtures/risky-diff.json new file mode 100644 index 000000000..3e97dccea --- /dev/null +++ b/skills/code-review-note/fixtures/risky-diff.json @@ -0,0 +1,9 @@ +{ + "pr_diff": "diff --git a/src/payments/refunds.ts b/src/payments/refunds.ts\nindex 1111111..2222222 100644\n--- a/src/payments/refunds.ts\n+++ b/src/payments/refunds.ts\n@@ -12,11 +12,11 @@ export async function approveRefund(req, user) {\n- if (!user.roles.includes(\"refund_admin\")) {\n- throw new Error(\"forbidden\");\n- }\n- if (req.amountCents > policy.maxAutoRefundCents) {\n- throw new Error(\"manual review required\");\n- }\n+ // TODO: restore role check after support migration\n+ const amount = Number(req.amountCents);\n+ if (!amount) return { status: \"skipped\" };\n return refundGateway.create({\n chargeId: req.chargeId,\n- amountCents: req.amountCents,\n+ amountCents: amount,\n });\n }\n", + "context": { + "repository": "example/payments", + "pr_number": "42", + "title": "Relax refund approval path", + "test_policy": "Require authorization and threshold regression tests for payment changes." + } +} diff --git a/skills/code-review-note/run.mjs b/skills/code-review-note/run.mjs new file mode 100644 index 000000000..de28c3176 --- /dev/null +++ b/skills/code-review-note/run.mjs @@ -0,0 +1,325 @@ +import fs from "node:fs"; + +const inputs = readInputs(); +const prDiff = stringValue(inputs.pr_diff); +const context = objectOrString(inputs.context); + +if (!prDiff || !hasDiffSignal(prDiff)) { + emit({ + schema: "code_review_note_packet.v1", + status: "refused", + refusal: { + reason_code: "invalid_input", + message: "pr_diff is empty or not a unified diff; refusing to invent review findings.", + }, + findings: [], + risk: { level: "unknown", rationale: "No parseable diff was supplied." }, + test_gaps: [], + review_note: null, + }); + process.exit(2); +} + +const parsed = parseDiff(prDiff); +if (parsed.files.length === 0) { + emit({ + schema: "code_review_note_packet.v1", + status: "refused", + refusal: { + reason_code: "invalid_input", + message: "No changed files were found in the supplied diff.", + }, + findings: [], + risk: { level: "unknown", rationale: "No changed files were supplied." }, + test_gaps: [], + review_note: null, + }); + process.exit(2); +} + +const findings = buildFindings(parsed); +const testGaps = buildTestGaps(parsed, findings, context); +const risk = summarizeRisk(findings, testGaps, parsed); +const reviewNote = buildReviewNote(findings, testGaps, risk, parsed, context); + +emit({ + schema: "code_review_note_packet.v1", + status: "sealed", + input: { + repository: field(context, "repository"), + pr_number: field(context, "pr_number"), + title: field(context, "title"), + diff_file_count: parsed.files.length, + }, + findings, + risk, + test_gaps: testGaps, + review_note: reviewNote, + guardrails: { + side_effects: "none", + posting_skill: "pr-review-note", + merge_scope: "refused", + grounding: "all findings cite changed files and visible diff lines", + }, +}); + +function readInputs() { + if (process.env.RUNX_INPUTS_PATH) { + return JSON.parse(fs.readFileSync(process.env.RUNX_INPUTS_PATH, "utf8")); + } + if (process.env.RUNX_INPUTS_JSON) { + return JSON.parse(process.env.RUNX_INPUTS_JSON); + } + return { + pr_diff: process.env.RUNX_INPUT_PR_DIFF, + context: parseMaybeJson(process.env.RUNX_INPUT_CONTEXT), + }; +} + +function parseMaybeJson(raw) { + if (raw === undefined || raw === "") return undefined; + try { + return JSON.parse(raw); + } catch { + return raw; + } +} + +function objectOrString(value) { + if (value && typeof value === "object" && !Array.isArray(value)) return value; + if (typeof value === "string") return { note: value }; + return {}; +} + +function stringValue(value) { + return typeof value === "string" ? value : null; +} + +function hasDiffSignal(diff) { + return diff.includes("diff --git ") || diff.includes("@@"); +} + +function parseDiff(diff) { + const files = []; + let current = null; + let oldLine = 0; + let newLine = 0; + + for (const line of diff.split(/\r?\n/)) { + const fileMatch = line.match(/^diff --git a\/(.+?) b\/(.+)$/); + if (fileMatch) { + current = { + old_path: fileMatch[1], + path: fileMatch[2], + hunks: [], + added: [], + removed: [], + }; + files.push(current); + continue; + } + if (!current) continue; + + const hunkMatch = line.match(/^@@ -(\d+)(?:,\d+)? \+(\d+)(?:,\d+)? @@/); + if (hunkMatch) { + oldLine = Number(hunkMatch[1]); + newLine = Number(hunkMatch[2]); + current.hunks.push(line); + continue; + } + if (line.startsWith("+") && !line.startsWith("+++")) { + current.added.push({ line: newLine, text: line.slice(1) }); + newLine += 1; + continue; + } + if (line.startsWith("-") && !line.startsWith("---")) { + current.removed.push({ line: oldLine, text: line.slice(1) }); + oldLine += 1; + continue; + } + if (!line.startsWith("\\ No newline")) { + oldLine += 1; + newLine += 1; + } + } + return { files }; +} + +function buildFindings(parsed) { + const findings = []; + for (const file of parsed.files) { + const addedText = file.added.map((line) => line.text).join("\n"); + const removedText = file.removed.map((line) => line.text).join("\n"); + const visibleText = `${addedText}\n${removedText}`; + const paymentFile = /pay|refund|invoice|charge|billing/i.test(file.path); + + if (paymentFile && removedMatches(removedText, ["role", "permission", "authorize", "forbidden", "admin"])) { + findings.push(finding({ + severity: "high", + category: "authorization_regression", + file, + evidenceLine: firstRemoved(file, ["role", "permission", "authorize", "forbidden", "admin"]), + summary: "Payment-adjacent code removes an authorization guard.", + reproduction: "Call the changed path as a non-privileged user and verify whether the request can reach the payment/refund gateway.", + })); + } + + if (paymentFile && removedMatches(removedText, ["max", "threshold", "limit", "manual review", "policy"])) { + findings.push(finding({ + severity: "high", + category: "policy_threshold_regression", + file, + evidenceLine: firstRemoved(file, ["max", "threshold", "limit", "manual review", "policy"]), + summary: "Payment policy threshold handling was removed or weakened.", + reproduction: "Submit an over-threshold payment/refund request and confirm it is escalated instead of proposed automatically.", + })); + } + + if (/\bTODO\b|restore|temporary|after migration/i.test(addedText)) { + findings.push(finding({ + severity: paymentFile ? "medium" : "low", + category: "temporary_control_gap", + file, + evidenceLine: firstAdded(file, ["TODO", "restore", "temporary", "migration"]), + summary: "New code marks a control as temporary or deferred.", + reproduction: "Inspect the follow-up condition and add a regression test that fails while the control remains disabled.", + })); + } + + if (/Number\(|parseInt\(|parseFloat\(/.test(addedText) && !/Number\.isFinite|isNaN|zod|schema|validate/i.test(visibleText)) { + findings.push(finding({ + severity: "medium", + category: "unchecked_numeric_parse", + file, + evidenceLine: firstAdded(file, ["Number(", "parseInt(", "parseFloat("]), + summary: "Numeric parsing is added without an explicit finite-number validation path.", + reproduction: "Exercise the path with empty, NaN, negative, and string amount inputs and verify the result cannot bypass policy checks.", + })); + } + } + return findings; +} + +function buildTestGaps(parsed, findings, context) { + const changedPaths = parsed.files.map((file) => file.path); + const hasTests = changedPaths.some((path) => /(^|\/)(test|tests|__tests__)\/|(\.|-)(test|spec)\.[jt]sx?$/i.test(path)); + const gaps = []; + + if (!hasTests && findings.some((item) => item.severity === "high")) { + gaps.push({ + name: "missing_high_risk_regression_tests", + detail: "No test file changed while high-risk behavior changed.", + expected_coverage: "Add authorization, policy-threshold, and negative-input regression tests for the changed path.", + }); + } + if (findings.some((item) => item.category === "unchecked_numeric_parse")) { + gaps.push({ + name: "missing_numeric_boundary_tests", + detail: "The diff adds numeric parsing without visible boundary coverage.", + expected_coverage: "Cover zero, empty string, NaN, negative, and over-limit values.", + }); + } + const policy = field(context, "test_policy"); + if (policy && gaps.length === 0 && !hasTests) { + gaps.push({ + name: "policy_named_tests_not_visible", + detail: `Context test policy says: ${policy}`, + expected_coverage: "Show the policy-specific regression test in the PR diff or link it in review context.", + }); + } + return gaps; +} + +function summarizeRisk(findings, testGaps, parsed) { + const severities = findings.map((item) => item.severity); + let level = "low"; + if (severities.includes("high")) level = "high"; + else if (severities.includes("medium")) level = "medium"; + if (level === "low" && testGaps.length > 0) level = "medium"; + return { + level, + finding_count: findings.length, + changed_files: parsed.files.map((file) => file.path), + rationale: findings.length > 0 + ? `Risk is ${level} because ${findings.length} grounded finding(s) were visible in the supplied diff.` + : "Risk is low because no concrete blocker was visible in the supplied diff.", + }; +} + +function buildReviewNote(findings, testGaps, risk, parsed, context) { + const repo = field(context, "repository") ?? "unknown repository"; + const pr = field(context, "pr_number") ?? "unknown PR"; + const lines = []; + lines.push(`Review note for ${repo}#${pr}`); + lines.push(""); + lines.push(`Risk: ${risk.level} (${risk.finding_count} finding(s))`); + lines.push(""); + if (findings.length > 0) { + lines.push("Findings:"); + for (const item of findings) { + lines.push(`- [${item.severity}] ${item.file}:${item.line} ${item.summary}`); + lines.push(` Reproduction: ${item.reproduction}`); + } + } else { + lines.push("Findings: no blocking issue visible in the supplied diff."); + } + lines.push(""); + if (testGaps.length > 0) { + lines.push("Test gaps:"); + for (const gap of testGaps) { + lines.push(`- ${gap.name}: ${gap.expected_coverage}`); + } + } else { + lines.push("Test gaps: none visible from the supplied context."); + } + lines.push(""); + lines.push("Scope: proposed comment only; posting requires pr.comment authority through pr-review-note. Merge is out of scope."); + + return { + effect: "proposed", + catalog_skill: "pr-review-note", + required_scope: "pr.comment", + merge_scope: "refused", + target: { repository: repo, pr_number: pr }, + body: lines.join("\n"), + }; +} + +function finding({ severity, category, file, evidenceLine, summary, reproduction }) { + return { + severity, + category, + file: file.path, + line: evidenceLine?.line ?? null, + evidence: evidenceLine?.text ?? null, + summary, + reproduction, + source: "supplied_diff", + }; +} + +function removedMatches(text, patterns) { + return patterns.some((pattern) => new RegExp(pattern, "i").test(text)); +} + +function firstRemoved(file, needles) { + return firstLine(file.removed, needles); +} + +function firstAdded(file, needles) { + return firstLine(file.added, needles); +} + +function firstLine(lines, needles) { + return lines.find((line) => needles.some((needle) => line.text.toLowerCase().includes(needle.toLowerCase()))) ?? lines[0] ?? null; +} + +function field(context, key) { + const value = context?.[key]; + if (typeof value === "string" && value.trim()) return value.trim(); + return null; +} + +function emit(value) { + process.stdout.write(`${JSON.stringify(value, null, 2)}\n`); +} diff --git a/skills/dependency-cve-audit/SKILL.md b/skills/dependency-cve-audit/SKILL.md new file mode 100644 index 000000000..1b2d634b6 --- /dev/null +++ b/skills/dependency-cve-audit/SKILL.md @@ -0,0 +1,185 @@ +--- +name: dependency-cve-audit +description: Audit locked npm dependencies against OSV advisories and emit exact-version CVE evidence. +source: + type: cli-tool + command: node + args: + - run.mjs +runx: + tags: + - security + - dependencies + - osv +links: + source: https://github.com/RYDE-PLAY/runx-dependency-cve-audit-skill + advisory_source: https://osv.dev +--- + +## What this skill does + +This skill audits a Node.js project's committed `package-lock.json` for known +vulnerabilities in npm dependencies. It extracts exact installed package +versions, queries the OSV API for each exact `{ ecosystem, package, version }` +tuple, and emits a machine-checkable evidence packet plus a concise Markdown +report. + +The skill is read-only. It does not install dependencies, execute target code, +modify repositories, open issues, submit advisories, or claim that a package is +safe when OSV returns no matching record. A zero-finding result means only that +OSV did not return advisories for the exact versions scanned under the selected +scope. + +## When to use this skill + +Use this skill when an agent needs a reproducible dependency vulnerability +snapshot for a public Node.js project, release candidate, benchmark fixture, or +security review packet. It is appropriate when the lockfile is public or has +been explicitly cleared for disclosure to OSV, and when the review needs exact +installed-version evidence rather than semver range guesses. + +The skill is especially useful before triage, upgrade planning, advisory +drafting, or reviewer handoff because it preserves the lockfile SHA-256, the +scan policy, the OSV query tuple for every finding, and references for each +advisory. + +## When not to use this skill + +Do not use this skill for private lockfiles unless the package names, versions, +and lockfile URL/path are approved for disclosure to the advisory source. Do not +use it as a full application security review, source-code audit, exploitability +assessment, SBOM generator, package installer, or automated remediation tool. + +Do not use the output as proof that a project is vulnerability-free. The result +depends on OSV coverage at run time, the selected dependency scope, and the +contents of the supplied lockfile. + +## Procedure + +1. Read `package_lock_path` or `package_lock_url`. +2. Record byte length and SHA-256 for the lockfile input. +3. Extract exact installed package versions from the lockfile. +4. Limit the scan to the declared `scan_scope` and `include_dev` policy. +5. Query OSV only with exact npm package names and exact installed versions. +6. Keep only vulnerabilities returned by OSV for that exact version query. +7. Emit `dependency.cve.audit.result.v1` with findings, query evidence, and validation. +8. Write `evidence.json` and `report.md` when `output_dir` is provided. + +## Edge cases and stop conditions + +Stop with an error when neither `package_lock_path` nor `package_lock_url` is +provided, when the lockfile cannot be read, when the URL is not HTTPS, when the +lockfile is not valid JSON, or when it does not contain a `packages` object. + +For local paths and output paths, the resolved path must stay inside the skill +directory. This prevents the runner from reading or writing unrelated workspace +files. + +If an OSV request fails, stop instead of returning a partial success packet. If +`scan_scope` is `direct`, skip direct dependencies whose installed package entry +is missing from `packages`; do not invent versions from semver declarations. + +The output is evidence for dependency triage, not an authorization to publish a +security advisory or mutate a repository. Any later issue filing, advisory +publication, or remediation PR needs its own gate. + +## Output schema + +The primary output is `dependency_cve_audit_result`, with schema +`dependency.cve.audit.result.v1`: + +```json +{ + "schema": "dependency.cve.audit.result.v1", + "data": { + "target": { + "name": "string | null", + "repo": "string | null", + "ref": "string | null" + }, + "source": { + "kind": "file | url", + "ref": "string", + "bytes": 0, + "sha256": "hex" + }, + "scanner": { + "name": "dependency-cve-audit", + "version": "0.1.1", + "advisory_source": "OSV.dev v1 query API", + "advisory_endpoint": "https://api.osv.dev/v1/query" + }, + "policy": { + "ecosystem": "npm", + "scan_scope": "direct | all", + "include_dev": false, + "target_code_executed": false, + "target_packages_installed": false, + "finding_rule": "string" + }, + "summary": { + "dependencies_scanned": 0, + "packages_with_findings": 0, + "findings": 0 + }, + "dependencies": [], + "findings": [], + "validation": { + "valid": true, + "every_finding_has_exact_version": true, + "every_finding_has_reference": true, + "every_finding_has_advisory_id": true, + "zero_false_hit_control": "string" + } + } +} +``` + +When `output_dir` is provided, the runner also writes `evidence.json` and +`report.md` inside that directory and returns their relative paths in +`data.artifacts`. + +## Worked example + +The harness scans OWASP NodeGoat at commit +`c5cb68a7084e4ae7dcc60e6a98768720a81841e8` using the committed +`package-lock.json`: + +```bash +runx skill "$PWD" \ + --input target_name="OWASP NodeGoat" \ + --input target_repo=https://github.com/OWASP/NodeGoat \ + --input target_ref=c5cb68a7084e4ae7dcc60e6a98768720a81841e8 \ + --input package_lock_url=https://raw.githubusercontent.com/OWASP/NodeGoat/c5cb68a7084e4ae7dcc60e6a98768720a81841e8/package-lock.json \ + --input scan_scope=direct \ + --input include_dev=false \ + --input output_dir=artifacts/nodegoat \ + --json +``` + +Expected evidence shape: + +- `source.sha256` records the fetched lockfile hash. +- `summary.dependencies_scanned` is the number of direct production npm + dependencies found in the lockfile. +- Each finding contains the exact installed version, OSV advisory id, aliases, + fixed versions when listed, affected ranges, and references. +- `validation.valid` is true only when every finding includes an exact version, + advisory id, and at least one reference. + +## Inputs + +- `target_name`: human-readable project name. +- `target_repo`: source repository URL. +- `target_ref`: immutable commit or release reference. +- `package_lock_path`: local path to a `package-lock.json`. +- `package_lock_url`: public URL for a `package-lock.json`. +- `scan_scope`: `direct` or `all`; defaults to `direct`. +- `include_dev`: include dev dependencies when true; defaults to false. +- `output_dir`: optional directory for `evidence.json` and `report.md`. + +## Outputs + +- `dependency_cve_audit_result`: complete packet. +- `evidence_json`: same evidence as machine-checkable JSON. +- `report_md`: concise report with exact version findings and references. diff --git a/skills/dependency-cve-audit/X.yaml b/skills/dependency-cve-audit/X.yaml new file mode 100644 index 000000000..bb3808809 --- /dev/null +++ b/skills/dependency-cve-audit/X.yaml @@ -0,0 +1,113 @@ +skill: dependency-cve-audit +version: "0.1.1" + +catalog: + kind: skill + audience: public + visibility: public + role: canonical + +runx: + mutating: false + idempotency: + key: lockfile_sha256 + scopes: + - dependencies.read + - advisories.osv.query + policy: + data_classification: public_dependency_metadata + network: + allowed: + - https://api.osv.dev/v1/query + - https://raw.githubusercontent.com/ + forbidden: + - private repositories + - package installation + - source code execution + verifier_notes: + - Every finding is produced from an exact package name and exact installed version query. + - The dogfood fixture pins the target repository to an immutable commit URL. + artifacts: + emits: + - dependency_cve_audit_result + - evidence_json + - report_md + wrap_as: dependency_cve_audit_packet + +harness: + cases: + - name: nodegoat-direct-production + runner: default + inputs: + target_name: OWASP NodeGoat + target_repo: https://github.com/OWASP/NodeGoat + target_ref: c5cb68a7084e4ae7dcc60e6a98768720a81841e8 + package_lock_url: https://raw.githubusercontent.com/OWASP/NodeGoat/c5cb68a7084e4ae7dcc60e6a98768720a81841e8/package-lock.json + scan_scope: direct + include_dev: false + output_dir: artifacts/nodegoat + expect: + status: sealed + +runners: + default: + default: true + type: cli-tool + command: /usr/bin/env + args: + - node + - run.mjs + scopes: + - dependencies.read + - advisories.osv.query + policy: + reads: + - public npm package lockfiles + calls: + - OSV exact-version query API + writes: + - evidence.json + - report.md + disallows: + - installing target packages + - executing target project code + - using private repository contents + inputs: + target_name: + type: string + required: true + description: Human-readable project name. + target_repo: + type: string + required: true + description: Public source repository URL. + target_ref: + type: string + required: false + description: Immutable commit, tag, or release reference. + package_lock_path: + type: string + required: false + description: Local package-lock.json path inside the skill directory. + package_lock_url: + type: string + required: false + description: Public package-lock.json URL. + scan_scope: + type: string + required: false + default: direct + description: direct or all dependencies. + include_dev: + type: boolean + required: false + default: false + description: Include development dependencies. + output_dir: + type: string + required: false + description: Directory inside the skill directory where artifacts should be written. + outputs: + dependency_cve_audit_result: object + evidence_json: object + report_md: string diff --git a/skills/dependency-cve-audit/run.mjs b/skills/dependency-cve-audit/run.mjs new file mode 100644 index 000000000..6a8eb140a --- /dev/null +++ b/skills/dependency-cve-audit/run.mjs @@ -0,0 +1,399 @@ +import crypto from "node:crypto"; +import fs from "node:fs"; +import path from "node:path"; + +const OSV_QUERY_URL = "https://api.osv.dev/v1/query"; +const SCHEMA = "dependency.cve.audit.result.v1"; + +const inputs = readInputs(); +const skillRoot = process.cwd(); +const scanScope = inputs.scan_scope || "direct"; +const includeDev = inputs.include_dev === true; + +if (!["direct", "all"].includes(scanScope)) { + throw new Error("scan_scope must be direct or all"); +} + +const source = await readLockfile(inputs, skillRoot); +const lockfile = JSON.parse(source.text); +const dependencies = collectDependencies(lockfile, { scanScope, includeDev }); +const findings = await queryFindings(dependencies); +const evidence = buildEvidence({ inputs, source, dependencies, findings, scanScope, includeDev }); +const report = renderReport(evidence); + +writeArtifacts(inputs.output_dir, evidence, report, skillRoot); + +process.stdout.write(`${JSON.stringify(evidence, null, 2)}\n`); + +function readInputs() { + const raw = process.env.RUNX_INPUTS_PATH + ? fs.readFileSync(process.env.RUNX_INPUTS_PATH, "utf8") + : process.env.RUNX_INPUTS_JSON || "{}"; + return JSON.parse(raw); +} + +async function readLockfile(rawInputs, root) { + if (typeof rawInputs.package_lock_path === "string" && rawInputs.package_lock_path.length > 0) { + const resolved = path.resolve(root, rawInputs.package_lock_path); + ensureInside(root, resolved, "package_lock_path"); + const text = fs.readFileSync(resolved, "utf8"); + return { kind: "file", ref: rawInputs.package_lock_path, text }; + } + if (typeof rawInputs.package_lock_url === "string" && rawInputs.package_lock_url.length > 0) { + const url = new URL(rawInputs.package_lock_url); + if (!["https:"].includes(url.protocol)) { + throw new Error("package_lock_url must be https"); + } + const response = await fetch(url, { headers: { accept: "application/json,text/plain,*/*" } }); + if (!response.ok) { + throw new Error(`GET ${url.href} returned ${response.status}`); + } + return { kind: "url", ref: url.href, text: await response.text() }; + } + throw new Error("package_lock_path or package_lock_url is required"); +} + +function collectDependencies(lockfile, { scanScope, includeDev }) { + if (!lockfile || typeof lockfile !== "object") { + throw new Error("package-lock.json must be a JSON object"); + } + if (!lockfile.packages || typeof lockfile.packages !== "object") { + throw new Error("package-lock.json packages object is required"); + } + + const root = lockfile.packages[""] || {}; + const prodDirect = new Set(Object.keys(root.dependencies || {})); + const devDirect = new Set(Object.keys(root.devDependencies || {})); + const directNames = new Set([...prodDirect, ...(includeDev ? devDirect : [])]); + const results = []; + + if (scanScope === "direct") { + for (const name of directNames) { + const pkgPath = `node_modules/${name}`; + const pkg = lockfile.packages[pkgPath]; + if (!pkg || typeof pkg !== "object" || typeof pkg.version !== "string") { + continue; + } + results.push(dependencyRecord({ + name, + pkg, + pkgPath, + prodDirect, + devDirect, + })); + } + return dedupeDependencies(results).sort((a, b) => a.name.localeCompare(b.name)); + } + + for (const [pkgPath, pkg] of Object.entries(lockfile.packages)) { + if (!pkgPath || !pkgPath.startsWith("node_modules/") || !pkg || typeof pkg !== "object") { + continue; + } + if (!pkg.version || typeof pkg.version !== "string") { + continue; + } + if (pkg.dev === true && !includeDev) { + continue; + } + + const name = packageNameFromLockPath(pkgPath); + results.push(dependencyRecord({ + name, + pkg, + pkgPath, + prodDirect, + devDirect, + })); + } + + return dedupeDependencies(results).sort((a, b) => a.name.localeCompare(b.name)); +} + +function packageNameFromLockPath(pkgPath) { + const marker = "node_modules/"; + const rest = pkgPath.slice(pkgPath.lastIndexOf(marker) + marker.length); + if (rest.startsWith("@")) { + const [scope, name] = rest.split("/"); + return `${scope}/${name}`; + } + return rest.split("/")[0]; +} + +function dependencyRecord({ name, pkg, pkgPath, prodDirect, devDirect }) { + const isProdDirect = prodDirect.has(name) && pkgPath === `node_modules/${name}`; + const isDevDirect = devDirect.has(name) && pkgPath === `node_modules/${name}`; + return { + ecosystem: "npm", + name, + version: pkg.version, + relation: isProdDirect ? "direct-production" : isDevDirect ? "direct-development" : "transitive", + lockfile_path: pkgPath, + resolved: typeof pkg.resolved === "string" ? pkg.resolved : null, + integrity: typeof pkg.integrity === "string" ? pkg.integrity : null, + }; +} + +function dedupeDependencies(dependencies) { + const seen = new Set(); + const results = []; + for (const dep of dependencies) { + const key = `${dep.name}@${dep.version}`; + if (!seen.has(key)) { + seen.add(key); + results.push(dep); + } + } + return results; +} + +async function queryFindings(dependencies) { + const findings = []; + for (const dependency of dependencies) { + const response = await fetch(OSV_QUERY_URL, { + method: "POST", + headers: { "content-type": "application/json", accept: "application/json" }, + body: JSON.stringify({ + version: dependency.version, + package: { + ecosystem: dependency.ecosystem, + name: dependency.name, + }, + }), + }); + if (!response.ok) { + throw new Error(`OSV query for ${dependency.name}@${dependency.version} returned ${response.status}`); + } + const payload = await response.json(); + for (const vuln of payload.vulns || []) { + findings.push(normalizeVulnerability(dependency, vuln)); + } + } + return findings.sort((a, b) => + `${a.dependency.name}:${a.advisory_id}`.localeCompare(`${b.dependency.name}:${b.advisory_id}`), + ); +} + +function normalizeVulnerability(dependency, vuln) { + const references = (vuln.references || []) + .map((ref) => ({ type: ref.type || "WEB", url: ref.url })) + .filter((ref) => typeof ref.url === "string" && ref.url.startsWith("http")); + const severities = (vuln.severity || []).map((entry) => `${entry.type}:${entry.score}`); + const aliases = Array.isArray(vuln.aliases) ? vuln.aliases : []; + + return { + dependency, + query: { + ecosystem: dependency.ecosystem, + package: dependency.name, + version: dependency.version, + advisory_source: OSV_QUERY_URL, + }, + advisory_id: vuln.id, + cve_ids: aliases.filter((alias) => alias.startsWith("CVE-")), + aliases, + summary: vuln.summary || "", + severity: severityLabel(vuln), + severity_vectors: severities, + fixed_versions: fixedVersions(vuln), + affected_ranges: affectedRangesForPackage(vuln, dependency.name), + published: vuln.published || null, + modified: vuln.modified || null, + references, + source_records: sourceRecords(vuln), + }; +} + +function severityLabel(vuln) { + const specific = vuln.database_specific || {}; + if (typeof specific.severity === "string" && specific.severity.length > 0) { + return specific.severity.toLowerCase(); + } + if (Array.isArray(vuln.severity) && vuln.severity.length > 0) { + return vuln.severity[0].score; + } + return "unknown"; +} + +function fixedVersions(vuln) { + const versions = new Set(); + for (const affected of vuln.affected || []) { + for (const range of affected.ranges || []) { + for (const event of range.events || []) { + if (event.fixed) versions.add(event.fixed); + } + } + } + return [...versions].sort(compareVersionish); +} + +function affectedRangesForPackage(vuln, packageName) { + const ranges = []; + for (const affected of vuln.affected || []) { + if (affected.package?.name !== packageName) continue; + for (const range of affected.ranges || []) { + ranges.push({ + type: range.type || null, + events: (range.events || []).map((event) => ({ ...event })), + }); + } + } + return ranges; +} + +function sourceRecords(vuln) { + const records = new Set(); + for (const affected of vuln.affected || []) { + const source = affected.database_specific?.source; + if (source) records.add(source); + } + return [...records].sort(); +} + +function compareVersionish(a, b) { + return String(a).localeCompare(String(b), undefined, { numeric: true, sensitivity: "base" }); +} + +function buildEvidence({ inputs, source, dependencies, findings, scanScope, includeDev }) { + const uniquePackagesWithFindings = new Set(findings.map((finding) => finding.dependency.name)); + const everyFindingHasExactVersion = findings.every((finding) => + finding.query.version === finding.dependency.version + && finding.query.package === finding.dependency.name + && finding.dependency.version.length > 0, + ); + const everyFindingHasReference = findings.every((finding) => finding.references.length > 0); + const everyFindingHasAdvisoryId = findings.every((finding) => finding.advisory_id.length > 0); + + return { + schema: SCHEMA, + data: { + target: { + name: inputs.target_name || null, + repo: inputs.target_repo || null, + ref: inputs.target_ref || null, + }, + source: { + kind: source.kind, + ref: source.ref, + bytes: Buffer.byteLength(source.text), + sha256: sha256(source.text), + }, + scanner: { + name: "dependency-cve-audit", + version: "0.1.1", + advisory_source: "OSV.dev v1 query API", + advisory_endpoint: OSV_QUERY_URL, + }, + policy: { + ecosystem: "npm", + scan_scope: scanScope, + include_dev: includeDev, + target_code_executed: false, + target_packages_installed: false, + finding_rule: "A finding is included only when OSV returns it for the exact npm package name and exact installed version from package-lock.json.", + }, + summary: { + dependencies_scanned: dependencies.length, + packages_with_findings: uniquePackagesWithFindings.size, + findings: findings.length, + }, + dependencies, + findings, + validation: { + valid: everyFindingHasExactVersion && everyFindingHasReference && everyFindingHasAdvisoryId, + every_finding_has_exact_version: everyFindingHasExactVersion, + every_finding_has_reference: everyFindingHasReference, + every_finding_has_advisory_id: everyFindingHasAdvisoryId, + zero_false_hit_control: "Each OSV request uses only the exact package name and version read from the lockfile; no inferred ranges or guessed versions are reported.", + }, + }, + }; +} + +function renderReport(packet) { + const data = packet.data; + const lines = []; + lines.push("# Dependency CVE Audit Report"); + lines.push(""); + lines.push(`Target: ${data.target.name}`); + lines.push(`Repository: ${data.target.repo}`); + lines.push(`Reference: ${data.target.ref}`); + lines.push(`Lockfile: ${data.source.ref}`); + lines.push(`Lockfile SHA-256: \`${data.source.sha256}\``); + lines.push(""); + lines.push("## Summary"); + lines.push(""); + lines.push(`- Advisory source: ${data.scanner.advisory_source} (${data.scanner.advisory_endpoint})`); + lines.push(`- Scan scope: ${data.policy.scan_scope} npm dependencies`); + lines.push(`- Include dev dependencies: ${data.policy.include_dev}`); + lines.push(`- Dependencies scanned: ${data.summary.dependencies_scanned}`); + lines.push(`- Packages with findings: ${data.summary.packages_with_findings}`); + lines.push(`- Exact-version findings: ${data.summary.findings}`); + lines.push(`- Target code executed: ${data.policy.target_code_executed}`); + lines.push(`- Target packages installed: ${data.policy.target_packages_installed}`); + lines.push(""); + lines.push("## Findings"); + lines.push(""); + + if (data.findings.length === 0) { + lines.push("No OSV vulnerabilities were returned for the scanned exact versions."); + } else { + lines.push("| Package | Version | Advisory | CVE aliases | Severity | Fixed versions | Primary reference |"); + lines.push("| --- | --- | --- | --- | --- | --- | --- |"); + for (const finding of data.findings) { + lines.push([ + finding.dependency.name, + finding.dependency.version, + finding.advisory_id, + finding.cve_ids.join(", ") || "none", + finding.severity, + finding.fixed_versions.join(", ") || "not listed", + finding.references[0]?.url || "not listed", + ].map(markdownCell).join(" | ").replace(/^/, "| ").replace(/$/, " |")); + } + } + + lines.push(""); + lines.push("## Reproducibility Controls"); + lines.push(""); + lines.push("- The lockfile URL is pinned to an immutable Git commit."); + lines.push("- Every dependency version comes from `package-lock.json`, not semver range resolution."); + lines.push("- Every finding is returned by OSV for an exact npm package and version query."); + lines.push("- The audit does not install packages or execute target project code."); + lines.push("- `evidence.json` contains the full dependency list, OSV query tuple, advisory IDs, aliases, references, and validation booleans."); + lines.push(""); + + return `${lines.join("\n")}\n`; +} + +function markdownCell(value) { + return String(value).replace(/\|/g, "\\|").replace(/\n/g, " "); +} + +function writeArtifacts(outputDir, evidence, report, root) { + if (!outputDir) { + evidence.data.artifacts = {}; + return; + } + const resolved = path.resolve(root, outputDir); + ensureInside(root, resolved, "output_dir"); + fs.mkdirSync(resolved, { recursive: true }); + const evidencePath = path.join(resolved, "evidence.json"); + const reportPath = path.join(resolved, "report.md"); + evidence.data.artifacts = { + evidence_json: path.relative(root, evidencePath), + report_md: path.relative(root, reportPath), + }; + fs.writeFileSync(evidencePath, `${JSON.stringify(evidence, null, 2)}\n`); + fs.writeFileSync(reportPath, report); +} + +function ensureInside(root, resolved, label) { + const normalizedRoot = root.endsWith(path.sep) ? root : `${root}${path.sep}`; + if (resolved !== root && !resolved.startsWith(normalizedRoot)) { + throw new Error(`${label} must stay inside the skill directory`); + } +} + +function sha256(value) { + return crypto.createHash("sha256").update(value).digest("hex"); +} diff --git a/skills/issue-intake/X.yaml b/skills/issue-intake/X.yaml index 617496a68..581a52a6b 100644 --- a/skills/issue-intake/X.yaml +++ b/skills/issue-intake/X.yaml @@ -73,6 +73,10 @@ harness: thread_locator: github://example/repo/issues/101 size: micro risk: low + closure: + disposition: closed + reason_code: issue_intake_completed + summary: Bounded docs fix is ready for the issue-to-pr lane. change_set: change_set_id: change_set_docs_work_101 thread_locator: github://example/repo/issues/101 @@ -205,6 +209,10 @@ harness: success_criteria: - One shared plan exists before repo mutation starts. - Repo-scoped workers receive explicit shared invariants. + closure: + disposition: closed + reason_code: issue_intake_completed + summary: Feature request needs a planning lane before mutation. change_set: change_set_id: change_set_abandoned_cart_982 thread_locator: support://request/982 @@ -292,6 +300,10 @@ harness: action_decision: stop review_target: none operator_notes: [] + closure: + disposition: closed + reason_code: issue_intake_completed + summary: Support question should be answered without a mutation lane. change_set: change_set_id: change_set_support_983 thread_locator: support://request/983 @@ -383,6 +395,10 @@ harness: target surface is explicit. operator_notes: - Do not open issue-to-pr until the target repo is explicit. + closure: + disposition: closed + reason_code: issue_intake_completed + summary: Ambiguous request needs review before mutation. change_set: change_set_id: change_set_support_984 thread_locator: github://example/repo/issues/984 diff --git a/skills/issue-to-pr/graph/scafld/run.mjs b/skills/issue-to-pr/graph/scafld/run.mjs index 39dc0900e..7a7ddb29e 100644 --- a/skills/issue-to-pr/graph/scafld/run.mjs +++ b/skills/issue-to-pr/graph/scafld/run.mjs @@ -12,6 +12,7 @@ const scafldSource = inputs.scafld_bin ? "env:SCAFLD_BIN" : "path:scafld"; const scafld = resolveBinary(scafldCandidate); +const minimumScafldVersion = String(inputs.scafld_min_version || "2.4.0"); const cwd = path.resolve(String( inputs.fixture || inputs.cwd @@ -136,6 +137,20 @@ if (path.isAbsolute(scafld) || scafld.includes(path.sep)) { env.PATH = `${path.dirname(scafld)}${path.delimiter}${env.PATH || "/usr/local/bin:/usr/bin:/bin"}`; } +try { + ensureScafldVersion({ + scafldBinary: scafld, + source: scafldSource, + requestedBinary: scafldCandidate, + workingDirectory: cwd, + processEnv: env, + minimum: minimumScafldVersion, + }); +} catch (error) { + console.error(error.message); + process.exit(1); +} + if (command === "build_to_review") { const outcome = runBuildToReview({ scafld, @@ -456,6 +471,62 @@ function parseMaxBuilds(value) { return 12; } +function ensureScafldVersion({ scafldBinary, source, requestedBinary, workingDirectory, processEnv, minimum }) { + const result = spawnSync(scafldBinary, ["--version"], { + cwd: workingDirectory, + env: processEnv, + encoding: "utf8", + shell: false, + }); + if (result.error) { + throw new Error(formatSpawnError({ + error: result.error, + source, + requestedBinary, + resolvedBinary: scafldBinary, + cwd: workingDirectory, + command: "--version", + args: ["--version"], + })); + } + + const exitCode = result.status ?? 1; + const rawVersion = `${result.stdout ?? ""}${result.stderr ?? ""}`.trim(); + if (exitCode !== 0) { + throw new Error(`scafld --version failed with exit ${exitCode}: ${rawVersion}`); + } + + const required = parseSemver(minimum); + if (!required) { + throw new Error(`invalid required scafld version: ${minimum}`); + } + + const actual = parseSemver(rawVersion); + if (!actual || compareSemver(actual, required) < 0) { + throw new Error( + `scafld ${minimum} or newer is required by this runx runner; ` + + `resolved ${scafldBinary} reported ${rawVersion || "no version"}`, + ); + } +} + +function parseSemver(value) { + const match = String(value).match(/\bv?(\d+)\.(\d+)\.(\d+)(?:[-+][0-9A-Za-z.-]+)?\b/); + if (!match) { + return null; + } + return [Number(match[1]), Number(match[2]), Number(match[3])]; +} + +function compareSemver(left, right) { + for (let index = 0; index < 3; index += 1) { + if (left[index] !== right[index]) { + return left[index] > right[index] ? 1 : -1; + } + } + return 0; +} + function firstNonEmptyString(...values) { for (const value of values) { if (typeof value !== "string") { diff --git a/skills/issue-to-pr/push-outbox/SKILL.md b/skills/issue-to-pr/push-outbox/SKILL.md index 89b720aeb..2d0dd16ca 100644 --- a/skills/issue-to-pr/push-outbox/SKILL.md +++ b/skills/issue-to-pr/push-outbox/SKILL.md @@ -1,6 +1,8 @@ --- name: issue-to-pr-push-outbox description: Publish issue-to-PR outbox entries through the governed Rust thread-outbox-provider front. +runx: + category: code source: type: thread-outbox-provider thread_outbox_provider: diff --git a/skills/least-privilege-auditor/X.yaml b/skills/least-privilege-auditor/X.yaml index 0c0fd88f4..019fc76db 100644 --- a/skills/least-privilege-auditor/X.yaml +++ b/skills/least-privilege-auditor/X.yaml @@ -5,12 +5,86 @@ catalog: audience: operator visibility: public role: canonical +harness: + cases: + - name: unused-scope-attenuation-proposed + inputs: + subject: skills/report-exporter + granted_scopes: + - drive.files.read:/reports/* + - drive.files.write:/reports/* + - drive.files.delete:/reports/* + usage_summary: + receipt_ids: + - rx_101 + - rx_102 + observed: + - scope: drive.files.read:/reports/* + count: 8 + refs: + - rx_101:step_3 + - rx_102:step_2 + - scope: drive.files.write:/reports/* + count: 2 + refs: + - rx_101:step_6 + - rx_102:step_5 + objective: Prepare the skill for renewal by removing unused authority. + caller: + answers: + agent_task.least-privilege-auditor.output: + audit_report: + status: attenuation_proposed + subject: skills/report-exporter + evidence: + receipt_ids: + - rx_101 + - rx_102 + receipt_window: null + grant_source: null + limitations: [] + removed_scopes: + - drive.files.delete:/reports/* + narrowed_scopes: [] + kept_scopes: + - drive.files.read:/reports/* + - drive.files.write:/reports/* + deferred_scopes: [] + attenuated_grant: + - drive.files.read:/reports/* + - drive.files.write:/reports/* + residual_risk: + - The subject can still read and write under /reports/*. + reviewer_action: applyable_now + attenuation_proposals: + - action: remove + scope: drive.files.delete:/reports/* + rationale: No cited receipt exercised delete authority. + verdict: "over-privileged: remove drive.files.delete:/reports/*" + expect: + status: sealed + receipt: + schema: runx.receipt.v1 + state: sealed + disposition: closed + + - name: missing-granted-scopes-needs-agent + inputs: + subject: skills/report-exporter + usage_summary: + receipt_ids: + - rx_101 + observed: [] + expect: + status: failure + runners: audit: default: true - type: agent-task - agent: reviewer - task: least-privilege-auditor + type: cli-tool + command: node + args: + - run.mjs outputs: audit_report: object attenuation_proposals: array diff --git a/skills/least-privilege-auditor/run.mjs b/skills/least-privilege-auditor/run.mjs new file mode 100644 index 000000000..1a4da7c74 --- /dev/null +++ b/skills/least-privilege-auditor/run.mjs @@ -0,0 +1,233 @@ +import fs from "node:fs"; + +const inputs = readInputs(); +const subject = stringValue(inputs.subject) || "unknown"; +const grantedScopes = stringArray(inputs.granted_scopes, "granted_scopes"); +const usageSummary = readUsageSummary(inputs.usage_summary); +const observed = collectObservedUsage(usageSummary); + +const scopeDiff = grantedScopes.map((scope) => classifyScope(scope, observed)); +const removedScopes = scopeDiff.filter((entry) => entry.classification === "remove").map((entry) => entry.granted_scope); +const narrowedScopes = scopeDiff + .filter((entry) => entry.classification === "narrow" && entry.proposal) + .map((entry) => ({ from: entry.granted_scope, to: entry.proposal })); +const keptScopes = scopeDiff.filter((entry) => entry.classification === "keep").map((entry) => entry.granted_scope); +const deferredScopes = scopeDiff.filter((entry) => entry.classification === "defer").map((entry) => entry.granted_scope); +const attenuatedGrant = [ + ...keptScopes, + ...narrowedScopes.map((entry) => entry.to), + ...deferredScopes, +]; + +const limitations = []; +if (observed.size === 0) { + limitations.push("No observed scope usage was provided; the grant cannot be safely narrowed."); +} + +const status = observed.size === 0 + ? "needs_more_evidence" + : removedScopes.length > 0 || narrowedScopes.length > 0 + ? "attenuation_proposed" + : "no_change"; + +const packet = { + status, + subject, + evidence: { + receipt_ids: Array.isArray(usageSummary.receipt_ids) ? usageSummary.receipt_ids.map(String) : [], + receipt_window: stringValue(usageSummary.receipt_window) || null, + grant_source: stringValue(inputs.grant_source) || null, + limitations, + }, + scope_diff: scopeDiff, + attenuated_grant: attenuatedGrant, + removed_scopes: removedScopes, + narrowed_scopes: narrowedScopes, + kept_scopes: keptScopes, + deferred_scopes: deferredScopes, + residual_risk: residualRisk({ keptScopes, deferredScopes, limitations }), + reviewer_action: status === "attenuation_proposed" + ? "applyable_now" + : status === "needs_more_evidence" + ? "gather_more_receipts" + : "none", + receipt_expectations: { + classification_counts: countClassifications(scopeDiff), + stop_status: status, + unresolved_questions: limitations, + }, +}; + +const result = { + audit_report: packet, + attenuation_proposals: [ + ...removedScopes.map((scope) => ({ + action: "remove", + scope, + rationale: "No cited receipt exercised this authority.", + })), + ...narrowedScopes.map((entry) => ({ + action: "narrow", + from: entry.from, + to: entry.to, + rationale: "Observed use fits the narrower grant.", + })), + ], + verdict: renderVerdict(packet), +}; + +process.stdout.write(`${JSON.stringify(result, null, 2)}\n`); + +function readInputs() { + const raw = process.env.RUNX_INPUTS_PATH + ? fs.readFileSync(process.env.RUNX_INPUTS_PATH, "utf8") + : process.env.RUNX_INPUTS_JSON || "{}"; + return JSON.parse(raw); +} + +function readUsageSummary(value) { + if (!value || typeof value !== "object" || Array.isArray(value)) { + throw new Error("usage_summary must be an object with receipt_ids and observed usage"); + } + return value; +} + +function stringArray(value, field) { + if (!Array.isArray(value) || value.length === 0) { + throw new Error(`${field} must be a non-empty array`); + } + return value.map((entry) => { + if (typeof entry !== "string" || entry.trim().length === 0) { + throw new Error(`${field} entries must be non-empty strings`); + } + return entry.trim(); + }); +} + +function collectObservedUsage(summary) { + const observed = new Map(); + const entries = Array.isArray(summary.observed) ? summary.observed : []; + for (const entry of entries) { + if (!entry || typeof entry !== "object") continue; + const scope = stringValue(entry.scope); + if (!scope) continue; + const current = observed.get(scope) || { count: 0, refs: [] }; + current.count += Number.isFinite(entry.count) ? Math.max(0, Math.trunc(entry.count)) : 1; + if (Array.isArray(entry.refs)) current.refs.push(...entry.refs.map(String)); + observed.set(scope, current); + } + return observed; +} + +function classifyScope(scope, observed) { + const normalized = normalizeScope(scope); + const exact = observed.get(scope); + if (exact && exact.count > 0) { + return diffEntry({ + scope, + normalized, + observedUse: { count: exact.count, receipt_refs: exact.refs }, + classification: "keep", + proposal: null, + rationale: "Observed receipt usage exercised this exact authority.", + }); + } + + const narrower = observedNarrowerScope(scope, observed); + if (narrower) { + return diffEntry({ + scope, + normalized, + observedUse: { count: narrower.count, receipt_refs: narrower.refs, scopes: narrower.scopes }, + classification: "narrow", + proposal: commonScopePrefix(narrower.scopes) || narrower.scopes[0], + rationale: "Observed usage fits a narrower scope than the granted wildcard.", + }); + } + + return diffEntry({ + scope, + normalized, + observedUse: { count: 0, receipt_refs: [] }, + classification: "remove", + proposal: null, + rationale: "No cited receipt exercised this authority.", + }); +} + +function observedNarrowerScope(scope, observed) { + if (!scope.endsWith("*")) return null; + const prefix = scope.slice(0, -1); + const matches = [...observed.entries()].filter(([used]) => used.startsWith(prefix)); + if (matches.length === 0) return null; + return { + scopes: matches.map(([used]) => used), + count: matches.reduce((sum, [, usage]) => sum + usage.count, 0), + refs: matches.flatMap(([, usage]) => usage.refs), + }; +} + +function normalizeScope(scope) { + const [verbPart, ...resourceParts] = scope.split(":"); + const resource = resourceParts.join(":") || null; + return { + verb: verbPart || null, + resource, + conditions: null, + }; +} + +function diffEntry({ scope, normalized, observedUse, classification, proposal, rationale }) { + return { + granted_scope: scope, + normalized, + observed_use: { + count: observedUse.count, + verbs: normalized.verb ? [normalized.verb] : [], + resources: normalized.resource ? [normalized.resource] : [], + receipt_refs: observedUse.receipt_refs || [], + scopes: observedUse.scopes || [], + }, + classification, + proposal, + rationale, + }; +} + +function commonScopePrefix(scopes) { + if (scopes.length !== 1) return null; + return scopes[0]; +} + +function countClassifications(entries) { + return entries.reduce((counts, entry) => { + counts[entry.classification] = (counts[entry.classification] || 0) + 1; + return counts; + }, {}); +} + +function residualRisk({ keptScopes, deferredScopes, limitations }) { + const risks = []; + if (keptScopes.length > 0) { + risks.push(`The subject still retains ${keptScopes.length} observed scope(s).`); + } + if (deferredScopes.length > 0) { + risks.push(`The subject has ${deferredScopes.length} deferred scope(s) requiring policy review.`); + } + risks.push(...limitations); + return risks; +} + +function renderVerdict(packet) { + if (packet.status === "attenuation_proposed") { + return `over-privileged: remove ${packet.removed_scopes.length}, narrow ${packet.narrowed_scopes.length}`; + } + if (packet.status === "needs_more_evidence") { + return "needs_more_evidence: no exercised scopes were provided"; + } + return "no_change: observed usage matches the grant"; +} + +function stringValue(value) { + return typeof value === "string" && value.trim().length > 0 ? value.trim() : null; +} diff --git a/skills/pr-review-note/SKILL.md b/skills/pr-review-note/SKILL.md index 135c2d0ca..2d3d7db6e 100644 --- a/skills/pr-review-note/SKILL.md +++ b/skills/pr-review-note/SKILL.md @@ -1,6 +1,8 @@ --- name: pr-review-note description: Govern a GitHub PR review-note lane over MCP; comment scope is admitted, merge scope is refused. +runx: + category: code --- # PR Review Note diff --git a/skills/review-skill/SKILL.md b/skills/review-skill/SKILL.md index aa230a7a0..3aef41f7a 100644 --- a/skills/review-skill/SKILL.md +++ b/skills/review-skill/SKILL.md @@ -31,9 +31,34 @@ adopt, publish, sandbox, or reject the skill. - Strategic bar: explain whether the skill strengthens the catalog, fills a real operator need, duplicates existing capability, or carries unacceptable trust risk. +- Public value bar: a skill is not publication-ready merely because it parses or + runs once. It should solve a real operator or user problem, be something the + catalog would stand behind, and produce evidence a stranger can verify. A + wrapper, placeholder, toy, or copied example with no credible adoption path is + a reject or sandbox-only recommendation. - Stop conditions: return `needs_more_evidence` when receipts or harness proof are missing, and `reject` when the skill cannot be bounded or audited. +## Review Gates + +Check these before recommending adoption or publication: + +- The `SKILL.md` states a bounded capability and does not promise more than the + execution profile implements. +- The execution profile declares typed inputs, outputs, side-effect posture, + allowed refs/tools, authority or approval posture, receipt mapping when a + domain act occurs, and harness cases. +- At least one meaningful happy path and one error or stop path are covered by + harness evidence or receipts. Local assertions without captured output are not + enough. +- Any published URL, registry listing, docs site, or repo is durable and public. + Placeholder hosts, private previews, unrelated parent domains, and dead links + lower trust or block publication. +- The evidence pack contains no secrets, private tokens, customer data, private + inbox content, or provider dumps. +- The recommendation states who would use or trust the skill and why. If that + answer is weak, recommend rejection, sandboxing, or a narrower redesign. + ## Output - `capability_profile`: what the skill appears to do and how it executes. diff --git a/skills/runx-operator/SKILL.md b/skills/runx-operator/SKILL.md index 0627dbf9b..cfa3e6197 100644 --- a/skills/runx-operator/SKILL.md +++ b/skills/runx-operator/SKILL.md @@ -105,6 +105,10 @@ product gap. Do not invent a private workaround. - Treat missing evidence as missing. Do not infer success from UI state alone. - Separate health, money, communications, provider mutations, access, deployment, and incident signals. + - For review, catalog, publication, bounty, or marketplace work, classify + whether the artifact is real, useful, complete, and valuable. A reachable + artifact with no credible user, maintainer, operator, public proof, or + marketing value is not ready. 3. Route to governed lanes. - Release questions route to `release` plus the project release profile and @@ -126,6 +130,9 @@ product gap. Do not invent a private workaround. - Live sends, payouts, refunds, customer-visible posts, provider mutations, target changes, credential changes, deploys, destructive actions, and broad audience decisions: explicit approval required. + - A review verdict, recommendation, or green dry-run is not payment approval. + Money movement needs a separate approval prompt naming the amount, recipient, + rail, target class, and verification receipt expected after settlement. - Missing approval means `awaiting_approval`, not "ready". 5. Produce the operator packet. @@ -236,6 +243,8 @@ operator_packet: provider response dumps. - Never claim a state is settled, sent, deployed, paid, or refunded without a receipt/effect/readback reference. +- Never route a public artifact, skill, bounty result, or docs deployment as + ready when it lacks a credible real-world audience or durable public evidence. - Never widen authority because a dashboard widget would be convenient. - Never duplicate an existing CLI command, workflow, hosted endpoint, or domain skill in operator prose. Route to it. diff --git a/skills/skill-testing/SKILL.md b/skills/skill-testing/SKILL.md index 49af0f562..0019a34cf 100644 --- a/skills/skill-testing/SKILL.md +++ b/skills/skill-testing/SKILL.md @@ -26,9 +26,30 @@ packages the approved output for publication or operator handoff. gaps directly. - Strategic bar: make adoption, sandboxing, rejection, or further testing easier. +- Public value bar: test whether the skill has a credible user, operator, + maintainer, or catalog reason to exist. Passing harnesses do not rescue a + placeholder, toy, duplicate, or low-value package. - Stop conditions: stop at review when trust evidence is insufficient or the skill cannot be bounded. +## Trust Audit Checks + +Before packaging a recommendation, confirm: + +- The skill contract is bounded and matches the execution profile. +- The execution profile declares typed inputs and outputs, side-effect posture, + allowed refs/tools, approval or authority posture, receipt mapping where + relevant, and harness cases. +- Harness or receipt evidence covers a meaningful happy path and at least one + stop or error path. +- Published artifacts are durable and public. Private previews, localhost, + placeholder hosts, unrelated parent domains, or dead links block a publication + recommendation. +- The audit names the concrete user-visible value: who would use, link, install, + trust, or maintain the skill. +- The evidence pack contains no secrets, raw credentials, private customer data, + private email bodies, wallet private keys, or provider response dumps. + ## Inputs - `skill_ref` (required): skill package or registry reference to assess. diff --git a/skills/sourcey/SKILL.md b/skills/sourcey/SKILL.md index a5368c89a..22f6c3241 100644 --- a/skills/sourcey/SKILL.md +++ b/skills/sourcey/SKILL.md @@ -69,6 +69,10 @@ would stand behind: demo framing unless the project itself uses that framing - never describe pages as machine output, agent output, or AI-generated docs; the site should read like the project maintainer wrote and stands behind it +- when publishing public docs, use a credible durable project, maintainer, + organization, product, or documentation home. Random personal domains, + placeholder parent sites, sandbox hosts, preview deploys, throwaway + subdomains, and unrelated novelty domains are not publication-quality homes - if the repo evidence is too thin for a strong docs page, surface that as an evidence gap instead of manufacturing confident filler @@ -88,6 +92,9 @@ would stand behind: - Strategic bar: the docs should make a real user action easier: install, evaluate, integrate, operate, or contribute. A pretty site with thin content is a failed run. +- Public value bar: a public Sourcey site should be something a real maintainer, + user, or ecosystem account would link. If the target project, host, or content + has no credible audience, return `needs_review` instead of shipping. - Stop conditions: return `needs_more_evidence`, `needs_review`, or an empty author/revise bundle when the repo already has the right docs or the evidence does not support new pages. @@ -223,6 +230,10 @@ with `needs_more_evidence` or `needs_review` instead of producing filler. - CI or deploy may run deterministic `sourcey build` from committed source. - Deploy must not be the step where docs scope, prose, or IA is invented. Do discovery, authoring, and review before deploy. +- For public publication, include enough proof for an external reviewer to + inspect the target project, source commit, Sourcey config or input source, + generated page list, deployment URL, parent domain, and durability of the + hosting choice. ## Config reference @@ -357,5 +368,8 @@ Invalid card icon names are a blocking quality issue. The build report includes missing ignore rule as an operational gap. - Build output may be regenerated in CI or deploy, but deploy must not author or revise docs content. +- Public deployments must be durable and socially credible. Do not treat a + throwaway preview URL, unrelated personal domain, placeholder parent site, or + sandbox subdomain as a completed public docs home. - Do not encode open-ended critique or revision behavior. Critique is one bounded evaluation pass. Revision is at most one explicit bounded pass. diff --git a/skills/spend/graph/pay-fulfill-rail/X.yaml b/skills/spend/graph/pay-fulfill-rail/X.yaml index 34655707a..2876ad7f2 100644 --- a/skills/spend/graph/pay-fulfill-rail/X.yaml +++ b/skills/spend/graph/pay-fulfill-rail/X.yaml @@ -91,16 +91,16 @@ runners: max_attempts: 1 idempotency: key: pay-fulfill-rail-mock - outputs: &a1 + outputs: rail_result: object rail_proof: object credential_envelope: object redactions: array recovery_hint: object - artifacts: &a2 + artifacts: wrap_as: effect_evidence_packet packet: runx.effect.evidence.v1 - runx: &a3 + runx: payment_authority: phase: fulfill resource_family: effect @@ -110,7 +110,7 @@ runners: authorization_form: single_use_capability receives_funding_material: false receipt_before_success: true - inputs: &a4 + inputs: payment_challenge: type: object required: true @@ -148,10 +148,16 @@ runners: max_attempts: 1 idempotency: key: payment-rail-x402 - outputs: *a1 - artifacts: *a2 + outputs: + rail_result: object + rail_proof: object + credential_envelope: object + redactions: array + recovery_hint: object + artifacts: + wrap_as: effect_evidence_packet + packet: runx.effect.evidence.v1 runx: - <<: *a3 payment_authority: phase: fulfill resource_family: effect @@ -163,7 +169,35 @@ runners: authorization_form: single_use_capability receives_funding_material: false receipt_before_success: true - inputs: *a4 + inputs: + payment_challenge: + type: object + required: true + description: Protocol/provider challenge to fulfill. + reserved_payment_authority: + type: object + required: true + description: Child payment authority term admitted by core. + spend_capability_ref: + type: object + required: true + description: Scoped single-use spend capability reference. + rail_profile_ref: + type: string + required: true + description: Configured rail profile reference. + payment_admission: + type: object + required: false + description: Hosted payment admission token and settlement identity. + idempotency: + type: object + required: true + description: Reservation key and recovery lookup fields. + quote_packet: + type: object + required: false + description: Source quote packet for evidence continuity. mpp: type: agent-task agent: operator @@ -173,10 +207,16 @@ runners: max_attempts: 1 idempotency: key: payment-rail-mpp - outputs: *a1 - artifacts: *a2 + outputs: + rail_result: object + rail_proof: object + credential_envelope: object + redactions: array + recovery_hint: object + artifacts: + wrap_as: effect_evidence_packet + packet: runx.effect.evidence.v1 runx: - <<: *a3 payment_authority: phase: fulfill resource_family: effect @@ -188,7 +228,35 @@ runners: authorization_form: single_use_capability receives_funding_material: false receipt_before_success: true - inputs: *a4 + inputs: + payment_challenge: + type: object + required: true + description: Protocol/provider challenge to fulfill. + reserved_payment_authority: + type: object + required: true + description: Child payment authority term admitted by core. + spend_capability_ref: + type: object + required: true + description: Scoped single-use spend capability reference. + rail_profile_ref: + type: string + required: true + description: Configured rail profile reference. + payment_admission: + type: object + required: false + description: Hosted payment admission token and settlement identity. + idempotency: + type: object + required: true + description: Reservation key and recovery lookup fields. + quote_packet: + type: object + required: false + description: Source quote packet for evidence continuity. stripe-spt: source: type: external-adapter @@ -199,10 +267,16 @@ runners: max_attempts: 1 idempotency: key: payment-rail-stripe-spt - outputs: *a1 - artifacts: *a2 + outputs: + rail_result: object + rail_proof: object + credential_envelope: object + redactions: array + recovery_hint: object + artifacts: + wrap_as: effect_evidence_packet + packet: runx.effect.evidence.v1 runx: - <<: *a3 payment_authority: phase: fulfill resource_family: effect @@ -214,4 +288,32 @@ runners: authorization_form: single_use_capability receives_funding_material: false receipt_before_success: true - inputs: *a4 + inputs: + payment_challenge: + type: object + required: true + description: Protocol/provider challenge to fulfill. + reserved_payment_authority: + type: object + required: true + description: Child payment authority term admitted by core. + spend_capability_ref: + type: object + required: true + description: Scoped single-use spend capability reference. + rail_profile_ref: + type: string + required: true + description: Configured rail profile reference. + payment_admission: + type: object + required: false + description: Hosted payment admission token and settlement identity. + idempotency: + type: object + required: true + description: Reservation key and recovery lookup fields. + quote_packet: + type: object + required: false + description: Source quote packet for evidence continuity. diff --git a/skills/structured-extraction/README.md b/skills/structured-extraction/README.md new file mode 100644 index 000000000..8e6dc7c9d --- /dev/null +++ b/skills/structured-extraction/README.md @@ -0,0 +1,21 @@ +# Structured Extraction Skill + +This skill extracts schema-validated JSON from messy HTML or text fixtures. The +default harness uses the real RFC 9110 HTML document as a deterministic input. + +It emits `runx.structured_extraction.result.v1` and includes artifact +references for: + +- input fixture SHA-256 +- JSON Schema SHA-256 +- validated output payload SHA-256 + +Reproduce: + +```powershell +runx harness . --receipt-dir .\receipts --json +``` + +The Frantic #22 delivery evidence was generated from +`fixtures/rfc9110-http-semantics.html` with source URL +`https://www.rfc-editor.org/rfc/rfc9110.html`. diff --git a/skills/structured-extraction/SKILL.md b/skills/structured-extraction/SKILL.md new file mode 100644 index 000000000..b2dc81b51 --- /dev/null +++ b/skills/structured-extraction/SKILL.md @@ -0,0 +1,34 @@ +--- +name: structured-extraction +description: Extract schema-validated JSON from messy HTML or text fixtures with digest-bound provenance. +source: + type: cli-tool + command: node + args: + - tools/structured/extract/run.mjs +runx: + tags: + - extraction + - schema-validation + - provenance +--- + +# Structured Extraction + +Use this skill to turn messy HTML or text into schema-validated JSON with +reproducible input and output digests. + +The default harness extracts a compact API-reference summary from the RFC 9110 +HTML document. It records the source URL, fixture byte count, input digest, +schema digest, extracted items, validation status, and artifact ids that the +runx receipt can bind as references. + +Inputs: + +- `input_path`: package-relative path to an HTML or text fixture. +- `schema_path`: package-relative JSON Schema path. +- `source_url`: canonical public source URL for the fixture. +- `content_type`: `text/html` or `text/plain`. +- `max_items`: maximum extracted items to include. + +The output packet is `runx.structured_extraction.result.v1`. diff --git a/skills/structured-extraction/X.yaml b/skills/structured-extraction/X.yaml new file mode 100644 index 000000000..0ff54732a --- /dev/null +++ b/skills/structured-extraction/X.yaml @@ -0,0 +1,83 @@ +skill: structured-extraction +version: "0.1.0" + +catalog: + kind: skill + audience: public + visibility: public + role: canonical + runtime_path: local + +policy: + allow: + - provider: local-fixture + method: READ + scope: runx:fixture:read + - provider: local-schema + method: READ + scope: runx:schema:read + deny: + - network_mutation + - credential_material + - private_user_data + +emits: + - name: structured_extraction_result + packet: runx.structured_extraction.result.v1 + +harness: + cases: + - name: rfc9110-http-semantics + runner: extract + inputs: + input_path: "fixtures/rfc9110-http-semantics.html" + schema_path: "schemas/extraction.schema.json" + source_url: "https://www.rfc-editor.org/rfc/rfc9110.html" + content_type: "text/html" + max_items: 18 + expect: + status: sealed + outputs: + structured_extraction_result: + matches_packet: runx.structured_extraction.result.v1 + receipt: + schema: runx.receipt.v1 + +runners: + extract: + default: true + type: graph + inputs: + input_path: + type: string + required: true + description: "Package-relative messy HTML or text fixture." + schema_path: + type: string + required: true + description: "Package-relative JSON Schema used to validate output." + source_url: + type: string + required: true + description: "Canonical source URL for the fixture bytes." + content_type: + type: string + required: false + default: "text/html" + description: "Input content type: text/html or text/plain." + max_items: + type: number + required: false + default: 20 + description: "Maximum extracted items to include in output." + graph: + name: structured-extraction + steps: + - id: extract_structured_json + tool: structured.extract + inputs: + input_path: "$input.input_path" + schema_path: "$input.schema_path" + source_url: "$input.source_url" + content_type: "$input.content_type" + max_items: "$input.max_items" diff --git a/skills/structured-extraction/fixtures/rfc9110-http-semantics.html b/skills/structured-extraction/fixtures/rfc9110-http-semantics.html new file mode 100644 index 000000000..10fbf5da9 --- /dev/null +++ b/skills/structured-extraction/fixtures/rfc9110-http-semantics.html @@ -0,0 +1,19015 @@ + + + + + + +RFC 9110: HTTP Semantics + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +
RFC 9110HTTP SemanticsJune 2022
Fielding, et al.Standards Track[Page]
+

+
+
+
Stream:
+
Internet Engineering Task Force (IETF)
+
RFC:
+
9110
+
STD:
+
97
+
Obsoletes:
+
+2818, 7230, 7231, 7232, 7233, 7235, 7538, 7615, 7694
+
Updates:
+
+3864
+
Category:
+
Standards Track
+
Published:
+
+ +
+
ISSN:
+
2070-1721
+
Authors:
+
+
+
R. Fielding, Ed. +
+
Adobe
+
+
+
M. Nottingham, Ed. +
+
Fastly
+
+
+
J. Reschke, Ed. +
+
greenbytes
+
+
+
+
+

RFC 9110

+

HTTP Semantics

+
+

Abstract

+

+ The Hypertext Transfer Protocol (HTTP) is a stateless application-level + protocol for distributed, collaborative, hypertext information systems. + This document describes the overall architecture of HTTP, establishes common + terminology, and defines aspects of the protocol that are shared by all + versions. In this definition are core protocol elements, extensibility + mechanisms, and the "http" and "https" Uniform Resource Identifier (URI) + schemes.¶

+

+ This document updates RFC 3864 and + obsoletes RFCs 2818, 7231, 7232, 7233, + 7235, 7538, 7615, 7694, and portions of 7230.¶

+
+
+
+

+Status of This Memo +

+

+ This is an Internet Standards Track document.¶

+

+ This document is a product of the Internet Engineering Task Force + (IETF). It represents the consensus of the IETF community. It has + received public review and has been approved for publication by + the Internet Engineering Steering Group (IESG). Further + information on Internet Standards is available in Section 2 of + RFC 7841.¶

+

+ Information about the current status of this document, any + errata, and how to provide feedback on it may be obtained at + https://www.rfc-editor.org/info/rfc9110.¶

+
+
+ +
+
+ ▲

+Table of Contents +

+ +
+
+
+
+

+1. Introduction +

+
+
+

+1.1. Purpose +

+

+ The Hypertext Transfer Protocol (HTTP) is a family of stateless, + application-level, request/response protocols that share a generic interface, + extensible semantics, and self-descriptive messages to enable flexible + interaction with network-based hypertext information systems.¶

+

+ HTTP hides the details of how a service is implemented by presenting a + uniform interface to clients that is independent of the types of resources + provided. Likewise, servers do not need to be aware of each client's + purpose: a request can be considered in isolation rather than being + associated with a specific type of client or a predetermined sequence of + application steps. This allows general-purpose implementations to be used + effectively in many different contexts, reduces interaction complexity, and + enables independent evolution over time.¶

+

+ HTTP is also designed for use as an intermediation protocol, wherein + proxies and gateways can translate non-HTTP information systems into a + more generic interface.¶

+

+ One consequence of this flexibility is that the protocol cannot be + defined in terms of what occurs behind the interface. Instead, we + are limited to defining the syntax of communication, the intent + of received communication, and the expected behavior of recipients. + If the communication is considered in isolation, then successful + actions ought to be reflected in corresponding changes to the + observable interface provided by servers. However, since multiple + clients might act in parallel and perhaps at cross-purposes, we + cannot require that such changes be observable beyond the scope + of a single response.¶

+
+
+
+
+

+1.2. History and Evolution +

+

+ HTTP has been the primary information transfer protocol for the World + Wide Web since its introduction in 1990. It began as a trivial + mechanism for low-latency requests, with a single method (GET) to + request transfer of a presumed hypertext document identified by a given pathname. + As the Web grew, HTTP was extended to enclose requests and responses within + messages, transfer arbitrary data formats using MIME-like media types, and + route requests through intermediaries. These protocols were eventually + defined as HTTP/0.9 and HTTP/1.0 (see [HTTP/1.0]).¶

+

+ HTTP/1.1 was designed to refine the protocol's features while retaining + compatibility with the existing text-based messaging syntax, improving + its interoperability, scalability, and robustness across the Internet. + This included length-based data delimiters for both fixed and dynamic + (chunked) content, a consistent framework for content negotiation, + opaque validators for conditional requests, cache controls for better + cache consistency, range requests for partial updates, and default + persistent connections. HTTP/1.1 was introduced in 1995 and published on + the Standards Track in 1997 [RFC2068], revised in + 1999 [RFC2616], and revised again in 2014 + ([RFC7230] through [RFC7235]).¶

+

+ HTTP/2 ([HTTP/2]) introduced a multiplexed session layer + on top of the existing TLS and TCP protocols for exchanging concurrent + HTTP messages with efficient field compression and server push. + HTTP/3 ([HTTP/3]) provides greater independence for concurrent + messages by using QUIC as a secure multiplexed transport over UDP instead of + TCP.¶

+

+ All three major versions of HTTP rely on the semantics defined by + this document. They have not obsoleted each other because each one has + specific benefits and limitations depending on the context of use. + Implementations are expected to choose the most appropriate transport and + messaging syntax for their particular context.¶

+

+ This revision of HTTP separates the definition of semantics (this document) + and caching ([CACHING]) from the current HTTP/1.1 messaging + syntax ([HTTP/1.1]) to allow each major protocol version + to progress independently while referring to the same core semantics.¶

+
+
+
+
+

+1.3. Core Semantics +

+

+ HTTP provides a uniform interface for interacting with a resource + (Section 3.1) -- regardless of its type, nature, or + implementation -- by sending messages that manipulate or transfer + representations (Section 3.2).¶

+

+ Each message is either a request or a response. A client constructs request + messages that communicate its intentions and routes those messages toward + an identified origin server. A server listens for requests, parses each + message received, interprets the message semantics in relation to the + identified target resource, and responds to that request with one or more + response messages. The client examines received responses to see if its + intentions were carried out, determining what to do next based on the + status codes and content received.¶

+

+ HTTP semantics include the intentions defined by each request method + (Section 9), extensions to those semantics that might be + described in request header fields, + status codes that describe the response (Section 15), and + other control data and resource metadata that might be given in response + fields.¶

+

+ + Semantics also include representation metadata that describe how + content is intended to be interpreted by a recipient, request header + fields that might influence content selection, and the various selection + algorithms that are collectively referred to as + "content negotiation" (Section 12).¶

+
+
+
+
+

+1.4. Specifications Obsoleted by This Document +

+ + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +
Table 1
TitleReferenceSee
HTTP Over TLS + [RFC2818] + + B.1 +
HTTP/1.1 Message Syntax and Routing [*] + [RFC7230] + + B.2 +
HTTP/1.1 Semantics and Content + [RFC7231] + + B.3 +
HTTP/1.1 Conditional Requests + [RFC7232] + + B.4 +
HTTP/1.1 Range Requests + [RFC7233] + + B.5 +
HTTP/1.1 Authentication + [RFC7235] + + B.6 +
HTTP Status Code 308 (Permanent Redirect) + [RFC7538] + + B.7 +
HTTP Authentication-Info and Proxy-Authentication-Info + Response Header Fields + [RFC7615] + + B.8 +
HTTP Client-Initiated Content-Encoding + [RFC7694] + + B.9 +
+

+ [*] This document only obsoletes the portions of + RFC 7230 that are independent of + the HTTP/1.1 messaging syntax and connection management; the remaining + bits of RFC 7230 are + obsoleted by "HTTP/1.1" [HTTP/1.1].¶

+
+
+
+
+
+
+

+2. Conformance +

+
+
+

+2.1. Syntax Notation +

+ + + + + + + + + + + + +

+ This specification uses the Augmented Backus-Naur Form (ABNF) notation of + [RFC5234], extended with the notation for case-sensitivity + in strings defined in [RFC7405].¶

+

+ It also uses a list extension, defined in Section 5.6.1, + that allows for compact definition of comma-separated lists using a "#" + operator (similar to how the "*" operator indicates repetition). Appendix A shows the collected grammar with all list + operators expanded to standard ABNF notation.¶

+

+ As a convention, ABNF rule names prefixed with "obs-" denote + obsolete grammar rules that appear for historical reasons.¶

+
+

+ + + + + + + + + + + + + The following core rules are included by + reference, as defined in Appendix B.1 of [RFC5234]: + ALPHA (letters), CR (carriage return), CRLF (CR LF), CTL (controls), + DIGIT (decimal 0-9), DQUOTE (double quote), + HEXDIG (hexadecimal 0-9/A-F/a-f), HTAB (horizontal tab), LF (line feed), + OCTET (any 8-bit sequence of data), SP (space), and + VCHAR (any visible US-ASCII character).¶

+
+

+ Section 5.6 defines some generic syntactic + components for field values.¶

+

+ This specification uses the terms + "character", + "character encoding scheme", + "charset", and + "protocol element" + as they are defined in [RFC6365].¶

+
+
+
+
+

+2.2. Requirements Notation +

+

+ The key words "MUST", "MUST NOT", + "REQUIRED", "SHALL", "SHALL NOT", + "SHOULD", "SHOULD NOT", + "RECOMMENDED", "NOT RECOMMENDED", + "MAY", and "OPTIONAL" in this document are to be + interpreted as described in BCP 14 [RFC2119] [RFC8174] when, and only when, they appear in all capitals, as + shown here.¶

+

+ This specification targets conformance criteria according to the role of + a participant in HTTP communication. Hence, requirements are placed + on senders, recipients, clients, servers, user agents, intermediaries, + origin servers, proxies, gateways, or caches, depending on what behavior + is being constrained by the requirement. Additional requirements + are placed on implementations, resource owners, and protocol element + registrations when they apply beyond the scope of a single communication.¶

+

+ The verb "generate" is used instead of "send" where a requirement applies + only to implementations that create the protocol element, rather than an + implementation that forwards a received element downstream.¶

+

+ An implementation is considered conformant if it complies with all of the + requirements associated with the roles it partakes in HTTP.¶

+

+ A sender MUST NOT generate protocol elements that do not match the grammar + defined by the corresponding ABNF rules. + Within a given message, a sender MUST NOT generate protocol elements or + syntax alternatives that are only allowed to be generated by participants in + other roles (i.e., a role that the sender does not have for that message).¶

+

+ Conformance to HTTP includes both conformance to the particular messaging + syntax of the protocol version in use and conformance to the semantics of + protocol elements sent. For example, a client that claims conformance to + HTTP/1.1 but fails to recognize the features required of HTTP/1.1 + recipients will fail to interoperate with servers that adjust their + responses in accordance with those claims. + Features that reflect user choices, such as content negotiation and + user-selected extensions, can impact application behavior beyond the + protocol stream; sending protocol elements that inaccurately reflect a + user's choices will confuse the user and inhibit choice.¶

+

+ When an implementation fails semantic conformance, recipients of that + implementation's messages will eventually develop workarounds to adjust + their behavior accordingly. A recipient MAY employ such workarounds while + remaining conformant to this protocol if the workarounds are limited to the + implementations at fault. For example, servers often scan portions of the + User-Agent field value, and user agents often scan the Server field value, + to adjust their own behavior with respect to known bugs or poorly chosen + defaults.¶

+
+
+
+
+

+2.3. Length Requirements +

+

+ A recipient SHOULD parse a received protocol element defensively, with + only marginal expectations that the element will conform to its ABNF + grammar and fit within a reasonable buffer size.¶

+

+ HTTP does not have specific length limitations for many of its protocol + elements because the lengths that might be appropriate will vary widely, + depending on the deployment context and purpose of the implementation. + Hence, interoperability between senders and recipients depends on shared + expectations regarding what is a reasonable length for each protocol + element. Furthermore, what is commonly understood to be a reasonable length + for some protocol elements has changed over the course of the past three + decades of HTTP use and is expected to continue changing in the future.¶

+

+ At a minimum, a recipient MUST be able to parse and process protocol + element lengths that are at least as long as the values that it generates + for those same protocol elements in other messages. For example, an origin + server that publishes very long URI references to its own resources needs + to be able to parse and process those same references when received as a + target URI.¶

+

+ Many received protocol elements are only parsed to the extent necessary to + identify and forward that element downstream. For example, an intermediary + might parse a received field into its field name and field value components, + but then forward the field without further parsing inside the field value.¶

+
+
+
+
+

+2.4. Error Handling +

+

+ A recipient MUST interpret a received protocol element according to the + semantics defined for it by this specification, including extensions to + this specification, unless the recipient has determined (through experience + or configuration) that the sender incorrectly implements what is implied by + those semantics. + For example, an origin server might disregard the contents of a received + Accept-Encoding header field if inspection of the + User-Agent header field indicates a specific implementation + version that is known to fail on receipt of certain content codings.¶

+

+ Unless noted otherwise, a recipient MAY attempt to recover a usable + protocol element from an invalid construct. HTTP does not define + specific error handling mechanisms except when they have a direct impact + on security, since different applications of the protocol require + different error handling strategies. For example, a Web browser might + wish to transparently recover from a response where the + Location header field doesn't parse according to the ABNF, + whereas a systems control client might consider any form of error recovery + to be dangerous.¶

+

+ Some requests can be automatically retried by a client in the event of + an underlying connection failure, as described in + Section 9.2.2.¶

+
+
+
+
+

+2.5. Protocol Version +

+

+ HTTP's version number consists of two decimal digits separated by a "." + (period or decimal point). The first digit (major version) indicates the + messaging syntax, whereas the second digit (minor version) + indicates the highest minor version within that major version to which the + sender is conformant (able to understand for future communication).¶

+

+ While HTTP's core semantics don't change between protocol versions, their + expression "on the wire" can change, and so the + HTTP version number changes when incompatible changes are made to the wire + format. Additionally, HTTP allows incremental, backwards-compatible + changes to be made to the protocol without changing its version through + the use of defined extension points (Section 16).¶

+

+ The protocol version as a whole indicates the sender's conformance with + the set of requirements laid out in that version's corresponding + specification(s). + For example, the version "HTTP/1.1" is defined by the combined + specifications of this document, "HTTP Caching" [CACHING], + and "HTTP/1.1" [HTTP/1.1].¶

+

+ HTTP's major version number is incremented when an incompatible message + syntax is introduced. The minor number is incremented when changes made to + the protocol have the effect of adding to the message semantics or + implying additional capabilities of the sender.¶

+

+ The minor version advertises the sender's communication capabilities even + when the sender is only using a backwards-compatible subset of the + protocol, thereby letting the recipient know that more advanced features + can be used in response (by servers) or in future requests (by clients).¶

+

+ When a major version of HTTP does not define any minor versions, the minor + version "0" is implied. The "0" is used when referring to that protocol + within elements that require a minor version identifier.¶

+
+
+
+
+
+
+

+3. Terminology and Core Concepts +

+

+ HTTP was created for the World Wide Web (WWW) architecture + and has evolved over time to support the scalability needs of a worldwide + hypertext system. Much of that architecture is reflected in the terminology + used to define HTTP.¶

+
+
+

+3.1. Resources +

+ +

+ The target of an HTTP request is called a "resource". + HTTP does not limit the nature of a resource; it merely + defines an interface that might be used to interact with resources. + Most resources are identified by a Uniform Resource Identifier (URI), as + described in Section 4.¶

+

+ One design goal of HTTP is to separate resource identification from + request semantics, which is made possible by vesting the request + semantics in the request method (Section 9) and a few + request-modifying header fields. + A resource cannot treat a request in a manner inconsistent with the + semantics of the method of the request. For example, though the URI of a + resource might imply semantics that are not safe, a client can expect the + resource to avoid actions that are unsafe when processing a request with a + safe method (see Section 9.2.1).¶

+

+ HTTP relies upon the Uniform Resource Identifier (URI) + standard [URI] to indicate the target resource + (Section 7.1) and relationships between resources.¶

+
+
+
+
+

+3.2. Representations +

+ +

+ A "representation" is information + that is intended to reflect a past, current, or desired state of a given + resource, in a format that can be readily communicated via the protocol. + A representation consists of a set of representation metadata and a + potentially unbounded stream of representation data + (Section 8).¶

+

+ HTTP allows "information hiding" behind its uniform interface by defining + communication with respect to a transferable representation of the resource + state, rather than transferring the resource itself. This allows the + resource identified by a URI to be anything, including temporal functions + like "the current weather in Laguna Beach", while potentially providing + information that represents that resource at the time a message is + generated [REST].¶

+

+ The uniform interface is similar to a window through which one can observe + and act upon a thing only through the communication of messages to an + independent actor on the other side. A shared abstraction is needed to + represent ("take the place of") the current or desired state of that thing + in our communications. When a representation is hypertext, it can provide + both a representation of the resource state and processing instructions + that help guide the recipient's future interactions.¶

+
+

+ + A target resource might be provided with, or be capable of + generating, multiple representations that are each intended to reflect the + resource's current state. An algorithm, usually based on + content negotiation (Section 12), + would be used to select one of those representations as being most + applicable to a given request. + This "selected representation" provides the data and metadata + for evaluating conditional requests (Section 13) + and constructing the content for 200 (OK), + 206 (Partial Content), and + 304 (Not Modified) responses to GET (Section 9.3.1).¶

+
+
+
+
+
+

+3.3. Connections, Clients, and Servers +

+ + + +

+ HTTP is a client/server protocol that operates over a reliable + transport- or session-layer "connection".¶

+

+ An HTTP "client" is a program that establishes a connection + to a server for the purpose of sending one or more HTTP requests. + An HTTP "server" is a program that accepts connections + in order to service HTTP requests by sending HTTP responses.¶

+

+ The terms client and server refer only to the roles that + these programs perform for a particular connection. The same program + might act as a client on some connections and a server on others.¶

+

+ HTTP is defined as a stateless protocol, meaning that each request message's semantics + can be understood in isolation, and that the relationship between connections + and messages on them has no impact on the interpretation of those messages. + For example, a CONNECT request (Section 9.3.6) or a request with + the Upgrade header field (Section 7.8) can occur at any time, + not just in the first message on a connection. Many implementations depend on + HTTP's stateless design in order to reuse proxied connections or dynamically + load balance requests across multiple servers.¶

+

+ As a result, a server MUST NOT + assume that two requests on the same connection are from the same user + agent unless the connection is secured and specific to that agent. + Some non-standard HTTP extensions (e.g., [RFC4559]) have + been known to violate this requirement, resulting in security and + interoperability problems.¶

+
+
+
+
+

+3.4. Messages +

+ + + + + + +

+ HTTP is a stateless request/response protocol for exchanging + "messages" across a connection. + The terms "sender" and "recipient" refer to + any implementation that sends or receives a given message, respectively.¶

+

+ A client sends requests to a server in the form of a "request" + message with a method (Section 9) and request target + (Section 7.1). The request might also contain + header fields (Section 6.3) for request modifiers, + client information, and representation metadata, + content (Section 6.4) intended for processing + in accordance with the method, and + trailer fields (Section 6.5) to communicate information + collected while sending the content.¶

+

+ A server responds to a client's request by sending one or more + "response" messages, each including a status + code (Section 15). The response might also contain + header fields for server information, resource metadata, and representation + metadata, content to be interpreted in accordance with the status + code, and trailer fields to communicate information + collected while sending the content.¶

+
+
+
+
+

+3.5. User Agents +

+ + + +

+ The term "user agent" refers to any of the various + client programs that initiate a request.¶

+

+ The most familiar form of user agent is the general-purpose Web browser, but + that's only a small percentage of implementations. Other common user agents + include spiders (web-traversing robots), command-line tools, billboard + screens, household appliances, scales, light bulbs, firmware update scripts, + mobile apps, and communication devices in a multitude of shapes and sizes.¶

+

+ Being a user agent does not imply that there is a human user directly + interacting with the software agent at the time of a request. In many + cases, a user agent is installed or configured to run in the background + and save its results for later inspection (or save only a subset of those + results that might be interesting or erroneous). Spiders, for example, are + typically given a start URI and configured to follow certain behavior while + crawling the Web as a hypertext graph.¶

+

+ Many user agents cannot, or choose not to, + make interactive suggestions to their user or provide adequate warning for + security or privacy concerns. In the few cases where this + specification requires reporting of errors to the user, it is acceptable + for such reporting to only be observable in an error console or log file. + Likewise, requirements that an automated action be confirmed by the user + before proceeding might be met via advance configuration choices, + run-time options, or simple avoidance of the unsafe action; confirmation + does not imply any specific user interface or interruption of normal + processing if the user has already made that choice.¶

+
+
+
+
+

+3.6. Origin Server +

+ +

+ The term "origin server" refers to a program that can + originate authoritative responses for a given target resource.¶

+

+ The most familiar form of origin server are large public websites. + However, like user agents being equated with browsers, it is easy to be + misled into thinking that all origin servers are alike. + Common origin servers also include home automation units, configurable + networking components, office machines, autonomous robots, news feeds, + traffic cameras, real-time ad selectors, and video-on-demand platforms.¶

+

+ Most HTTP communication consists of a retrieval request (GET) for + a representation of some resource identified by a URI. In the + simplest case, this might be accomplished via a single bidirectional + connection (===) between the user agent (UA) and the origin server (O).¶

+
+
+
+         request   >
+    UA ======================================= O
+                                <   response
+
+
+
Figure 1
+
+
+
+
+

+3.7. Intermediaries +

+ +

+ HTTP enables the use of intermediaries to satisfy requests through + a chain of connections. There are three common forms of HTTP + "intermediary": proxy, gateway, and tunnel. In some cases, + a single intermediary might act as an origin server, proxy, gateway, + or tunnel, switching behavior based on the nature of each request.¶

+
+
+
+         >             >             >             >
+    UA =========== A =========== B =========== C =========== O
+               <             <             <             <
+
+
+
Figure 2
+

+ The figure above shows three intermediaries (A, B, and C) between the + user agent and origin server. A request or response message that + travels the whole chain will pass through four separate connections. + Some HTTP communication options + might apply only to the connection with the nearest, non-tunnel + neighbor, only to the endpoints of the chain, or to all connections + along the chain. Although the diagram is linear, each participant might + be engaged in multiple, simultaneous communications. For example, B + might be receiving requests from many clients other than A, and/or + forwarding requests to servers other than C, at the same time that it + is handling A's request. Likewise, later requests might be sent through a + different path of connections, often based on dynamic configuration for + load balancing.¶

+

+ + + + + The terms "upstream" and "downstream" are + used to describe directional requirements in relation to the message flow: + all messages flow from upstream to downstream. + The terms "inbound" and "outbound" are used to describe directional + requirements in relation to the request route: + inbound means "toward the origin server", whereas + outbound means "toward the user agent".¶

+

+ + A "proxy" is a message-forwarding agent that is chosen by the + client, usually via local configuration rules, to receive requests + for some type(s) of absolute URI and attempt to satisfy those + requests via translation through the HTTP interface. Some translations + are minimal, such as for proxy requests for "http" URIs, whereas + other requests might require translation to and from entirely different + application-level protocols. Proxies are often used to group an + organization's HTTP requests through a common intermediary for the + sake of security services, annotation services, or shared caching. Some + proxies are designed to apply transformations to selected messages or + content while they are being forwarded, as described in + Section 7.7.¶

+

+ + + + A "gateway" (a.k.a. "reverse proxy") is an + intermediary that acts as an origin server for the outbound connection but + translates received requests and forwards them inbound to another server or + servers. Gateways are often used to encapsulate legacy or untrusted + information services, to improve server performance through + "accelerator" caching, and to enable partitioning or load + balancing of HTTP services across multiple machines.¶

+

+ All HTTP requirements applicable to an origin server + also apply to the outbound communication of a gateway. + A gateway communicates with inbound servers using any protocol that + it desires, including private extensions to HTTP that are outside + the scope of this specification. However, an HTTP-to-HTTP gateway + that wishes to interoperate with third-party HTTP servers needs to conform + to user agent requirements on the gateway's inbound connection.¶

+

+ + A "tunnel" acts as a blind relay between two connections + without changing the messages. Once active, a tunnel is not + considered a party to the HTTP communication, though the tunnel might + have been initiated by an HTTP request. A tunnel ceases to exist when + both ends of the relayed connection are closed. Tunnels are used to + extend a virtual connection through an intermediary, such as when + Transport Layer Security (TLS, [TLS13]) is used to + establish confidential communication through a shared firewall proxy.¶

+

+ The above categories for intermediary only consider those acting as + participants in the HTTP communication. There are also intermediaries + that can act on lower layers of the network protocol stack, filtering or + redirecting HTTP traffic without the knowledge or permission of message + senders. Network intermediaries are indistinguishable (at a protocol level) + from an on-path attacker, often introducing security flaws or + interoperability problems due to mistakenly violating HTTP semantics.¶

+

+ + + For example, an "interception proxy" [RFC3040] (also commonly + known as a "transparent proxy" [RFC1919]) + differs from an HTTP proxy because it is not chosen by the client. + Instead, an interception proxy filters or redirects outgoing TCP port 80 + packets (and occasionally other common port traffic). + Interception proxies are commonly found on public network access points, + as a means of enforcing account subscription prior to allowing use of + non-local Internet services, and within corporate firewalls to enforce + network usage policies.¶

+
+
+
+
+

+3.8. Caches +

+ +

+ A "cache" is a local store of previous response messages and the + subsystem that controls its message storage, retrieval, and deletion. + A cache stores cacheable responses in order to reduce the response + time and network bandwidth consumption on future, equivalent + requests. Any client or server MAY employ a cache, though a cache + cannot be used while acting as a tunnel.¶

+

+ The effect of a cache is that the request/response chain is shortened + if one of the participants along the chain has a cached response + applicable to that request. The following illustrates the resulting + chain if B has a cached copy of an earlier response from O (via C) + for a request that has not been cached by UA or A.¶

+
+
+
+            >             >
+       UA =========== A =========== B - - - - - - C - - - - - - O
+                  <             <
+
+
+
Figure 3
+

+ + A response is "cacheable" if a cache is allowed to store a copy of + the response message for use in answering subsequent requests. + Even when a response is cacheable, there might be additional + constraints placed by the client or by the origin server on when + that cached response can be used for a particular request. HTTP + requirements for cache behavior and cacheable responses are + defined in [CACHING].¶

+

+ There is a wide variety of architectures and configurations + of caches deployed across the World Wide Web and + inside large organizations. These include national hierarchies + of proxy caches to save bandwidth and reduce latency, content delivery + networks that use gateway caching to optimize regional and global distribution of popular sites, + collaborative systems that + broadcast or multicast cache entries, archives of pre-fetched cache + entries for use in off-line or high-latency environments, and so on.¶

+
+
+
+
+

+3.9. Example Message Exchange +

+

+ The following example illustrates a typical HTTP/1.1 message exchange for a + GET request (Section 9.3.1) on the URI "http://www.example.com/hello.txt":¶

+

+Client request:¶

+
+
GET /hello.txt HTTP/1.1
+User-Agent: curl/7.64.1
+Host: www.example.com
+Accept-Language: en, mi
+
+
¶ +
+

+Server response:¶

+
+
HTTP/1.1 200 OK
+Date: Mon, 27 Jul 2009 12:28:53 GMT
+Server: Apache
+Last-Modified: Wed, 22 Jul 2009 19:15:56 GMT
+ETag: "34aa387-d-1568eb00"
+Accept-Ranges: bytes
+Content-Length: 51
+Vary: Accept-Encoding
+Content-Type: text/plain
+
+Hello World! My content includes a trailing CRLF.
+
¶ +
+
+
+
+
+
+
+

+4. Identifiers in HTTP +

+ + +

+ Uniform Resource Identifiers (URIs) [URI] are used + throughout HTTP as the means for identifying resources (Section 3.1).¶

+
+
+

+4.1. URI References +

+ +

+ URI references are used to target requests, indicate redirects, and define + relationships.¶

+

+ The definitions of "URI-reference", + "absolute-URI", "relative-part", "authority", "port", "host", + "path-abempty", "segment", and "query" are adopted from the + URI generic syntax. + An "absolute-path" rule is defined for protocol elements that can contain a + non-empty path component. (This rule differs slightly from the path-abempty + rule of RFC 3986, which allows for an empty path, + and path-absolute rule, which does not allow paths that begin with "//".) + A "partial-URI" rule is defined for protocol elements + that can contain a relative URI but not a fragment component.¶

+ + + + + + + + + +
+
  URI-reference = <URI-reference, see [URI], Section 4.1>
+  absolute-URI  = <absolute-URI, see [URI], Section 4.3>
+  relative-part = <relative-part, see [URI], Section 4.2>
+  authority     = <authority, see [URI], Section 3.2>
+  uri-host      = <host, see [URI], Section 3.2.2>
+  port          = <port, see [URI], Section 3.2.3>
+  path-abempty  = <path-abempty, see [URI], Section 3.3>
+  segment       = <segment, see [URI], Section 3.3>
+  query         = <query, see [URI], Section 3.4>
+
+  absolute-path = 1*( "/" segment )
+  partial-URI   = relative-part [ "?" query ]
+
¶ +
+

+ Each protocol element in HTTP that allows a URI reference will indicate + in its ABNF production whether the element allows any form of reference + (URI-reference), only a URI in absolute form (absolute-URI), only the + path and optional query components (partial-URI), + or some combination of the above. + Unless otherwise indicated, URI references are parsed + relative to the target URI (Section 7.1).¶

+

+ It is RECOMMENDED that all senders and recipients support, at a minimum, + URIs with lengths of 8000 octets in protocol elements. Note that this + implies some structures and on-wire representations (for example, the + request line in HTTP/1.1) will necessarily be larger in some cases.¶

+
+
+
+
+ +

+ IANA maintains the registry of URI Schemes [BCP35] at + <https://www.iana.org/assignments/uri-schemes/>. + Although requests might target any URI scheme, the following schemes are + inherent to HTTP servers:¶

+
+ + + + + + + + + + + + + + + + + + + + + +
Table 2
URI SchemeDescriptionSection
httpHypertext Transfer Protocol + 4.2.1 +
httpsHypertext Transfer Protocol Secure + 4.2.2 +
+
+

+ Note that the presence of an "http" or "https" URI does not imply that + there is always an HTTP server at the identified origin listening for + connections. Anyone can mint a URI, whether or not a server exists and + whether or not that server currently maps that identifier to a resource. + The delegated nature of registered names and IP addresses creates a + federated namespace whether or not an HTTP server is present.¶

+
+
+

+4.2.1. http URI Scheme +

+ + +

+ The "http" URI scheme is hereby defined for minting identifiers within the + hierarchical namespace governed by a potential HTTP origin server + listening for TCP ([TCP]) connections on a given port.¶

+ +
+
  http-URI = "http" "://" authority path-abempty [ "?" query ]
+
¶ +
+

+ The origin server for an "http" URI is identified by the + authority component, which includes a host identifier + ([URI], Section 3.2.2) + and optional port number ([URI], Section 3.2.3). + If the port subcomponent is empty or not given, TCP port 80 (the + reserved port for WWW services) is the default. + The origin determines who has the right to respond authoritatively to + requests that target the identified resource, as defined in + Section 4.3.2.¶

+

+ A sender MUST NOT generate an "http" URI with an empty host identifier. + A recipient that processes such a URI reference MUST reject it as invalid.¶

+

+ The hierarchical path component and optional query component identify the + target resource within that origin server's namespace.¶

+
+
+
+
+

+4.2.2. https URI Scheme +

+ + + +

+ The "https" URI scheme is hereby defined for minting identifiers within the + hierarchical namespace governed by a potential origin server listening for + TCP connections on a given port and capable of establishing a TLS + ([TLS13]) connection that has been secured for HTTP + communication. In this context, "secured" specifically + means that the server has been authenticated as acting on behalf of the + identified authority and all HTTP communication with that server has + confidentiality and integrity protection that is acceptable to both client + and server.¶

+ +
+
  https-URI = "https" "://" authority path-abempty [ "?" query ]
+
¶ +
+

+ The origin server for an "https" URI is identified by the + authority component, which includes a host identifier + ([URI], Section 3.2.2) + and optional port number ([URI], Section 3.2.3). + If the port subcomponent is empty or not given, TCP port 443 + (the reserved port for HTTP over TLS) is the default. + The origin determines who has the right to respond authoritatively to + requests that target the identified resource, as defined in + Section 4.3.3.¶

+

+ A sender MUST NOT generate an "https" URI with an empty host identifier. + A recipient that processes such a URI reference MUST reject it as invalid.¶

+

+ The hierarchical path component and optional query component identify the + target resource within that origin server's namespace.¶

+

+ A client MUST ensure that its HTTP requests for an "https" resource are + secured, prior to being communicated, and that it only accepts secured + responses to those requests. Note that the definition of what cryptographic + mechanisms are acceptable to client and server are usually negotiated and + can change over time.¶

+

+ Resources made available via the "https" scheme have no shared identity + with the "http" scheme. They are distinct origins with separate namespaces. + However, extensions to HTTP that are defined as applying to all origins with + the same host, such as the Cookie protocol [COOKIE], + allow information set by one service to impact communication with other + services within a matching group of host domains. Such extensions ought to + be designed with great care to prevent information obtained from a secured + connection being inadvertently exchanged within an unsecured context.¶

+
+
+
+
+

+4.2.3. http(s) Normalization and Comparison +

+

+ URIs with an "http" or "https" scheme are normalized and compared according to the + methods defined in Section 6 of [URI], using + the defaults described above for each scheme.¶

+

+ HTTP does not require the use of a specific method for determining + equivalence. For example, a cache key might be compared as a simple + string, after syntax-based normalization, or after scheme-based + normalization.¶

+

+ Scheme-based normalization (Section 6.2.3 of [URI]) of "http" and "https" URIs involves the following + additional rules:¶

+
    +
  • If the port is equal to the default port for a scheme, the normal form + is to omit the port subcomponent.¶ +
    • +
  • When not being used as the target of an OPTIONS request, an empty path + component is equivalent to an absolute path of "/", so the normal form is + to provide a path of "/" instead.¶ +
    • +
  • The scheme and host are case-insensitive and normally provided in + lowercase; all other components are compared in a case-sensitive + manner.¶ +
    • +
  • Characters other than those in the "reserved" set are equivalent to + their percent-encoded octets: the normal form is to not encode them (see + Sections 2.1 and 2.2 of [URI]).¶ +
    • +
+

+ For example, the following three URIs are equivalent:¶

+
+
+   http://example.com:80/~smith/home.html
+   http://EXAMPLE.com/%7Esmith/home.html
+   http://EXAMPLE.com:/%7esmith/home.html
+
¶ +
+

+ Two HTTP URIs that are equivalent after normalization (using any method) + can be assumed to identify the same resource, and any HTTP component MAY + perform normalization. As a result, distinct resources SHOULD NOT be + identified by HTTP URIs that are equivalent after normalization (using any + method defined in Section 6.2 of [URI]).¶

+
+
+
+
+

+4.2.4. Deprecation of userinfo in http(s) URIs +

+

+ The URI generic syntax for authority also includes a userinfo subcomponent + ([URI], Section 3.2.1) for including user + authentication information in the URI. In that subcomponent, the + use of the format "user:password" is deprecated.¶

+

+ Some implementations make use of the userinfo component for internal + configuration of authentication information, such as within command + invocation options, configuration files, or bookmark lists, even + though such usage might expose a user identifier or password.¶

+

+ A sender MUST NOT generate the userinfo subcomponent (and its "@" + delimiter) when an "http" or "https" URI reference is generated within a + message as a target URI or field value.¶

+

+ Before making use of an "http" or "https" URI reference received from an untrusted + source, a recipient SHOULD parse for userinfo and treat its presence as + an error; it is likely being used to obscure the authority for the sake of + phishing attacks.¶

+
+
+
+
+

+4.2.5. http(s) References with Fragment Identifiers +

+ +

+ Fragment identifiers allow for indirect identification + of a secondary resource, independent of the URI scheme, as defined in + Section 3.5 of [URI]. + Some protocol elements that refer to a URI allow inclusion of a fragment, + while others do not. They are distinguished by use of the ABNF rule for + elements where fragment is allowed; otherwise, a specific rule that excludes + fragments is used.¶

+ +
+
+
+
+
+
+

+4.3. Authoritative Access +

+

+ Authoritative access refers to dereferencing a given identifier, + for the sake of access to the identified resource, in a way that the client + believes is authoritative (controlled by the resource owner). The process + for determining whether access is granted is defined by the URI scheme and often uses + data within the URI components, such as the authority component when + the generic syntax is used. However, authoritative access is not limited to + the identified mechanism.¶

+

+ Section 4.3.1 defines the concept of an origin as an aid to + such uses, and the subsequent subsections explain how to establish that a + peer has the authority to represent an origin.¶

+

+ See Section 17.1 for security considerations + related to establishing authority.¶

+
+
+

+4.3.1. URI Origin +

+ + +

+ The "origin" for a given URI is the triple of scheme, host, + and port after normalizing the scheme and host to lowercase and + normalizing the port to remove any leading zeros. If port is elided from + the URI, the default port for that scheme is used. For example, the URI¶

+
+
+   https://Example.Com/happy.js
+
¶ +
+

+ would have the origin¶

+
+
+   { "https", "example.com", "443" }
+
¶ +
+

+ which can also be described as the normalized URI prefix with port always + present:¶

+
+
+   https://example.com:443
+
¶ +
+

+ Each origin defines its own namespace and controls how identifiers + within that namespace are mapped to resources. In turn, how the origin + responds to valid requests, consistently over time, determines the + semantics that users will associate with a URI, and the usefulness of + those semantics is what ultimately transforms these mechanisms into a + resource for users to reference and access in the future.¶

+

+ Two origins are distinct if they differ in scheme, host, or port. Even + when it can be verified that the same entity controls two distinct origins, + the two namespaces under those origins are distinct unless explicitly + aliased by a server authoritative for that origin.¶

+

+ Origin is also used within HTML and related Web protocols, beyond the + scope of this document, as described in [RFC6454].¶

+
+
+
+
+

+4.3.2. http Origins +

+

+ Although HTTP is independent of the transport protocol, the "http" scheme + (Section 4.2.1) is specific to associating authority with + whomever controls the origin + server listening for TCP connections on the indicated port of whatever + host is identified within the authority component. This is a very weak + sense of authority because it depends on both client-specific name + resolution mechanisms and communication that might not be secured from + an on-path attacker. Nevertheless, it is a sufficient minimum for + binding "http" identifiers to an origin server for consistent resolution + within a trusted environment.¶

+

+ If the host identifier is provided as an IP address, the origin server is + the listener (if any) on the indicated TCP port at that IP address. + If host is a registered name, the registered name is an indirect identifier + for use with a name resolution service, such as DNS, to find an address for + an appropriate origin server.¶

+

+ When an "http" URI is used within a context that calls for access to the + indicated resource, a client MAY attempt access by resolving the host + identifier to an IP address, establishing a TCP connection to that + address on the indicated port, and sending over that connection an HTTP + request message containing a request target that matches the client's + target URI (Section 7.1).¶

+

+ If the server responds to such a request with a non-interim HTTP response + message, as described in Section 15, then that response + is considered an authoritative answer to the client's request.¶

+

+ Note, however, that the above is not the only means for obtaining an + authoritative response, nor does it imply that an authoritative response + is always necessary (see [CACHING]). + For example, the Alt-Svc header field [ALTSVC] allows an + origin server to identify other services that are also authoritative for + that origin. Access to "http" identified resources might also be provided + by protocols outside the scope of this document.¶

+
+
+
+
+

+4.3.3. https Origins +

+

+ The "https" scheme (Section 4.2.2) associates authority based + on the ability of a server to use the private key corresponding to a + certificate that the client considers to be trustworthy for the identified + origin server. The client usually relies upon a chain of trust, conveyed + from some prearranged or configured trust anchor, to deem a certificate + trustworthy (Section 4.3.4).¶

+

+ In HTTP/1.1 and earlier, a client will only attribute authority to a server + when they are communicating over a successfully established and secured + connection specifically to that URI origin's host. The connection + establishment and certificate verification are used as proof of authority.¶

+

+ In HTTP/2 and HTTP/3, a client will attribute authority to a server when + they are communicating over a successfully established and secured + connection if the URI origin's host matches any of the hosts present in the + server's certificate and the client believes that it could open a connection + to that host for that URI. In practice, a client will make a DNS query to + check that the origin's host contains the same server IP address as the + established connection. This restriction can be removed by the origin server + sending an equivalent ORIGIN frame [RFC8336].¶

+

+ The request target's host and port value are passed within each HTTP + request, identifying the origin and distinguishing it from other namespaces + that might be controlled by the same server (Section 7.2). + It is the origin's responsibility to ensure that any services provided with + control over its certificate's private key are equally responsible for + managing the corresponding "https" namespaces or at least prepared to + reject requests that appear to have been misdirected + (Section 7.4).¶

+

+ An origin server might be unwilling to process requests for certain target + URIs even when they have the authority to do so. For example, when a host + operates distinct services on different ports (e.g., 443 and 8000), checking + the target URI at the origin server is necessary (even after the connection + has been secured) because a network attacker might cause connections for one + port to be received at some other port. Failing to check the target URI + might allow such an attacker to replace a response to one target URI + (e.g., "https://example.com/foo") with a seemingly authoritative response + from the other port (e.g., "https://example.com:8000/foo").¶

+

+ Note that the "https" scheme does not rely on TCP and the connected port + number for associating authority, since both are outside the secured + communication and thus cannot be trusted as definitive. Hence, the HTTP + communication might take place over any channel that has been secured, + as defined in Section 4.2.2, including protocols that don't + use TCP.¶

+

+ When an "https" URI is used within a context that calls for access to + the indicated resource, a client MAY attempt access by resolving the + host identifier to an IP address, establishing a TCP connection to that + address on the indicated port, securing the connection end-to-end by + successfully initiating TLS over TCP with confidentiality and integrity + protection, and sending over that connection an HTTP request message + containing a request target that matches the client's target URI + (Section 7.1).¶

+

+ If the server responds to such a request with a non-interim HTTP response + message, as described in Section 15, then that response + is considered an authoritative answer to the client's request.¶

+

+ Note, however, that the above is not the only means for obtaining an + authoritative response, nor does it imply that an authoritative response + is always necessary (see [CACHING]).¶

+
+
+
+
+

+4.3.4. https Certificate Verification +

+

+ To establish a secured connection to dereference a URI, + a client MUST verify that the service's identity is an acceptable + match for the URI's origin server. Certificate verification is used to + prevent server impersonation by an on-path attacker or by an attacker + that controls name resolution. This process requires that a client be + configured with a set of trust anchors.¶

+

+ In general, a client MUST verify the service identity using the + verification process defined in + Section 6 of [RFC6125]. The client MUST construct + a reference identity from the service's host: if the host is a literal IP address + (Section 4.3.5), the reference identity is an IP-ID, otherwise + the host is a name and the reference identity is a DNS-ID.¶

+

+ A reference identity of type CN-ID MUST NOT be used by clients. As noted + in Section 6.2.1 of [RFC6125], a reference + identity of type CN-ID might be used by older clients.¶

+

+ A client might be specially configured to accept an alternative form of + server identity verification. For example, a client might be connecting + to a server whose address and hostname are dynamic, with an expectation that + the service will present a specific certificate (or a certificate matching + some externally defined reference identity) rather than one matching the + target URI's origin.¶

+

+ In special cases, it might be appropriate for + a client to simply ignore the server's identity, but it must be + understood that this leaves a connection open to active attack.¶

+

+ If the certificate is not valid for the target URI's origin, + a user agent MUST either obtain confirmation from the user + before proceeding (see Section 3.5) or + terminate the connection with a bad certificate error. Automated + clients MUST log the error to an appropriate audit log (if available) + and SHOULD terminate the connection (with a bad certificate error). + Automated clients MAY provide a configuration setting that disables + this check, but MUST provide a setting which enables it.¶

+
+
+
+
+

+4.3.5. IP-ID Reference Identity +

+

+ A server that is identified using an IP address literal in the "host" field + of an "https" URI has a reference identity of type IP-ID. An IP version 4 + address uses the "IPv4address" ABNF rule, and an IP version 6 address uses + the "IP-literal" production with the "IPv6address" option; see + Section 3.2.2 of [URI]. A reference identity of + IP-ID contains the decoded bytes of the IP address.¶

+

+ An IP version 4 address is 4 octets, and an IP version 6 address is 16 octets. + Use of IP-ID is not defined for any other IP version. The iPAddress + choice in the certificate subjectAltName extension does not explicitly + include the IP version and so relies on the length of the address to + distinguish versions; see + Section 4.2.1.6 of [RFC5280].¶

+

+ A reference identity of type IP-ID matches if the address is identical to + an iPAddress value of the subjectAltName extension of the certificate.¶

+
+
+
+
+
+
+
+
+

+5. Fields +

+ +

+ HTTP uses "fields" to provide data in the form of extensible + name/value pairs with a registered key namespace. Fields are sent and + received within the header and trailer sections of messages + (Section 6).¶

+
+
+

+5.1. Field Names +

+

+ A field name labels the corresponding field value as having the + semantics defined by that name. For example, the Date + header field is defined in Section 6.6.1 as containing the + origination timestamp for the message in which it appears.¶

+ +
+
  field-name     = token
+
¶ +
+

+ Field names are case-insensitive and ought to be registered within the + "Hypertext Transfer Protocol (HTTP) Field Name Registry"; see Section 16.3.1.¶

+

+ The interpretation of a field does not change between minor + versions of the same major HTTP version, though the default behavior of a + recipient in the absence of such a field can change. Unless specified + otherwise, fields are defined for all versions of HTTP. + In particular, the Host and Connection + fields ought to be recognized by all HTTP implementations + whether or not they advertise conformance with HTTP/1.1.¶

+

+ New fields can be introduced without changing the protocol version if + their defined semantics allow them to be safely ignored by recipients + that do not recognize them; see Section 16.3.¶

+

+ A proxy MUST forward unrecognized header fields unless the + field name is listed in the Connection header field + (Section 7.6.1) or the proxy is specifically + configured to block, or otherwise transform, such fields. + Other recipients SHOULD ignore unrecognized header and trailer fields. + Adhering to these requirements allows HTTP's functionality to be extended + without updating or removing deployed intermediaries.¶

+
+
+
+
+

+5.2. Field Lines and Combined Field Value +

+

+ + + + Field sections are composed of any number of "field lines", + each with a "field name" (see Section 5.1) + identifying the field, and a "field line value" that conveys + data for that instance of the field.¶

+

+ + When a field name is only present once in a section, the combined + "field value" for that field consists of the corresponding + field line value. + When a field name is repeated within a section, its combined field value + consists of the list of corresponding field line values within that section, + concatenated in order, with each field line value separated by a comma.¶

+

+ For example, this section:¶

+
+
Example-Field: Foo, Bar
+Example-Field: Baz
+
¶ +
+

+ contains two field lines, both with the field name "Example-Field". The + first field line has a field line value of "Foo, Bar", while the second + field line value is "Baz". The field value for "Example-Field" is the list + "Foo, Bar, Baz".¶

+
+
+
+
+

+5.3. Field Order +

+

+ A recipient MAY combine multiple field lines within a field section that + have the same field name + into one field line, without changing the semantics of the message, by + appending each subsequent field line value to the initial field line value + in order, separated by a comma (",") and optional whitespace + (OWS, defined in Section 5.6.3). + For consistency, use comma SP.¶

+

+ The order in which field lines with the + same name are received is therefore significant to the interpretation of + the field value; a proxy MUST NOT change the order of these field line + values when forwarding a message.¶

+

+ This means that, aside from the well-known exception noted below, a sender + MUST NOT generate multiple field lines with the same name in a message + (whether in the headers or trailers) or append a field line when a field + line of the same name already exists in the message, unless that field's + definition allows multiple field line values to be recombined as a + comma-separated list (i.e., at least one alternative of the field's + definition allows a comma-separated list, such as an ABNF rule of + #(values) defined in Section 5.6.1).¶

+ +

+ The order in which field lines with differing field names are received in a + section is not significant. However, it is good practice to send header + fields that contain additional control data first, such as + Host on requests and Date on responses, so + that implementations can decide when not to handle a message as early as + possible.¶

+

+ A server MUST NOT apply a request to the target resource until it + receives the entire request header section, since later header field lines + might include conditionals, authentication credentials, or deliberately + misleading duplicate header fields that could impact request processing.¶

+
+
+
+
+

+5.4. Field Limits +

+

+ HTTP does not place a predefined limit on the length of each field line, field value, + or on the length of a header or trailer section as a whole, as described in + Section 2. Various ad hoc limitations on individual + lengths are found in practice, often depending on the specific + field's semantics.¶

+

+ A server that receives a request header field line, field value, or set of + fields larger than it wishes to process MUST respond with an appropriate + 4xx (Client Error) status code. Ignoring such header fields + would increase the server's vulnerability to request smuggling attacks + (Section 11.2 of [HTTP/1.1]).¶

+

+ A client MAY discard or truncate received field lines that are larger + than the client wishes to process if the field semantics are such that the + dropped value(s) can be safely ignored without changing the + message framing or response semantics.¶

+
+
+
+
+

+5.5. Field Values +

+

+ HTTP field values consist of a sequence of characters in a format defined + by the field's grammar. Each field's grammar is usually defined using + ABNF ([RFC5234]).¶

+ + + + +
+
  field-value    = *field-content
+  field-content  = field-vchar
+                   [ 1*( SP / HTAB / field-vchar ) field-vchar ]
+  field-vchar    = VCHAR / obs-text
+  obs-text       = %x80-FF
+
¶ +
+

+ A field value does not include leading or trailing whitespace. When a + specific version of HTTP allows such whitespace to appear in a message, + a field parsing implementation MUST exclude such whitespace prior to + evaluating the field value.¶

+

+ Field values are usually constrained to the range of US-ASCII characters + [USASCII]. + Fields needing a greater range of characters can use an encoding, + such as the one defined in [RFC8187]. + Historically, HTTP allowed field content with text in the ISO-8859-1 + charset [ISO-8859-1], supporting other charsets only + through use of [RFC2047] encoding. + Specifications for newly defined fields SHOULD limit their values to + visible US-ASCII octets (VCHAR), SP, and HTAB. + A recipient SHOULD treat other allowed octets in field content + (i.e., obs-text) as opaque data.¶

+

+ Field values containing CR, LF, or NUL characters are invalid and dangerous, + due to the varying ways that implementations might parse and interpret + those characters; a recipient of CR, LF, or NUL within a field value MUST + either reject the message or replace each of those characters with SP + before further processing or forwarding of that message. Field values + containing other CTL characters are also invalid; however, + recipients MAY retain such characters for the sake of robustness when + they appear within a safe context (e.g., an application-specific quoted + string that will not be processed by any downstream HTTP parser).¶

+

+ + Fields that only anticipate a single member as the field value are + referred to as "singleton fields".¶

+

+ + Fields that allow multiple members as the field value are referred to as + "list-based fields". The list operator extension of + Section 5.6.1 is used as a common notation for defining + field values that can contain multiple members.¶

+

+ Because commas (",") are used as the delimiter between members, they need + to be treated with care if they are allowed as data within a member. This + is true for both list-based and singleton fields, since a singleton field + might be erroneously sent with multiple members and detecting such errors + improves interoperability. Fields that expect to contain a + comma within a member, such as within an HTTP-date or + URI-reference + element, ought to be defined with delimiters around that element to + distinguish any comma within that data from potential list separators.¶

+

+ For example, a textual date and a URI (either of which might contain a comma) + could be safely carried in list-based field values like these:¶

+
+
Example-URIs: "http://example.com/a.html,foo",
+              "http://without-a-comma.example.com/"
+Example-Dates: "Sat, 04 May 1996", "Wed, 14 Sep 2005"
+
¶ +
+

+ Note that double-quote delimiters are almost always used with the + quoted-string production (Section 5.6.4); using a different syntax inside double-quotes + will likely cause unnecessary confusion.¶

+

+ Many fields (such as Content-Type, defined in + Section 8.3) use a common syntax for parameters + that allows both unquoted (token) and quoted (quoted-string) syntax for + a parameter value (Section 5.6.6). Use of common syntax + allows recipients to reuse existing parser components. When allowing both + forms, the meaning of a parameter value ought to be the same whether it + was received as a token or a quoted string.¶

+ +
+
+
+
+

+5.6. Common Rules for Defining Field Values +

+
+
+

+5.6.1. Lists (#rule ABNF Extension) +

+

+ A #rule extension to the ABNF rules of [RFC5234] is used to + improve readability in the definitions of some list-based field values.¶

+

+ A construct "#" is defined, similar to "*", for defining comma-delimited + lists of elements. The full form is "<n>#<m>element" indicating + at least <n> and at most <m> elements, each separated by a single + comma (",") and optional whitespace (OWS, + defined in Section 5.6.3).¶

+
+
+
+5.6.1.1. Sender Requirements +
+

+ In any production that uses the list construct, a sender MUST NOT + generate empty list elements. In other words, a sender has to generate + lists that satisfy the following syntax:¶

+
+
+  1#element => element *( OWS "," OWS element )
+
¶ +
+

+ and:¶

+
+
+  #element => [ 1#element ]
+
¶ +
+

+ and for n >= 1 and m > 1:¶

+
+
+  <n>#<m>element => element <n-1>*<m-1>( OWS "," OWS element )
+
¶ +
+

+ Appendix A shows the collected ABNF for senders + after the list constructs have been expanded.¶

+
+
+
+
+
+5.6.1.2. Recipient Requirements +
+

+ Empty elements do not contribute to the count of elements present. + A recipient MUST parse and ignore + a reasonable number of empty list elements: enough to handle common mistakes + by senders that merge values, but not so much that they could be used as a + denial-of-service mechanism. In other words, a recipient MUST accept lists + that satisfy the following syntax:¶

+
+
+  #element => [ element ] *( OWS "," OWS [ element ] )
+
¶ +
+

+ Note that because of the potential presence of empty list elements, the + RFC 5234 ABNF cannot enforce the cardinality of list elements, and + consequently all cases are mapped as if there was no cardinality specified.¶

+

+ For example, given these ABNF productions:¶

+
+
+  example-list      = 1#example-list-elmt
+  example-list-elmt = token ; see Section 5.6.2
+
¶ +
+

+ Then the following are valid values for example-list (not including the + double quotes, which are present for delimitation only):¶

+
+
+  "foo,bar"
+  "foo ,bar,"
+  "foo , ,bar,charlie"
+
¶ +
+

+ In contrast, the following values would be invalid, since at least one + non-empty element is required by the example-list production:¶

+
+
+  ""
+  ","
+  ",   ,"
+
¶ +
+
+
+
+
+
+
+

+5.6.2. Tokens +

+
+

+ + + Tokens are short textual identifiers that do not include whitespace or + delimiters.¶

+
+ + +
+
  token          = 1*tchar
+
+  tchar          = "!" / "#" / "$" / "%" / "&" / "'" / "*"
+                 / "+" / "-" / "." / "^" / "_" / "`" / "|" / "~"
+                 / DIGIT / ALPHA
+                 ; any VCHAR, except delimiters
+
¶ +
+
+

+ + Many HTTP field values are defined using common syntax + components, separated by whitespace or specific delimiting characters. + Delimiters are chosen from the set of US-ASCII visual characters not + allowed in a token (DQUOTE and "(),/:;<=>?@[\]{}").¶

+
+
+
+
+
+

+5.6.3. Whitespace +

+

+ This specification uses three rules to denote the use of linear + whitespace: OWS (optional whitespace), RWS (required whitespace), and + BWS ("bad" whitespace).¶

+

+ The OWS rule is used where zero or more linear whitespace octets might + appear. For protocol elements where optional whitespace is preferred to + improve readability, a sender SHOULD generate the optional whitespace + as a single SP; otherwise, a sender SHOULD NOT generate optional + whitespace except as needed to overwrite invalid or unwanted protocol + elements during in-place message filtering.¶

+

+ The RWS rule is used when at least one linear whitespace octet is required + to separate field tokens. A sender SHOULD generate RWS as a single SP.¶

+

+ OWS and RWS have the same semantics as a single SP. Any content known to + be defined as OWS or RWS MAY be replaced with a single SP before + interpreting it or forwarding the message downstream.¶

+

+ The BWS rule is used where the grammar allows optional whitespace only for + historical reasons. A sender MUST NOT generate BWS in messages. + A recipient MUST parse for such bad whitespace and remove it before + interpreting the protocol element.¶

+

+ BWS has no semantics. Any content known to be + defined as BWS MAY be removed before interpreting it or forwarding the + message downstream.¶

+ + + +
+
  OWS            = *( SP / HTAB )
+                 ; optional whitespace
+  RWS            = 1*( SP / HTAB )
+                 ; required whitespace
+  BWS            = OWS
+                 ; "bad" whitespace
+
¶ +
+
+
+
+
+

+5.6.4. Quoted Strings +

+
+

+ + + A string of text is parsed as a single value if it is quoted using + double-quote marks.¶

+
+ + +
+
  quoted-string  = DQUOTE *( qdtext / quoted-pair ) DQUOTE
+  qdtext         = HTAB / SP / %x21 / %x23-5B / %x5D-7E / obs-text
+
¶ +
+
+

+ + The backslash octet ("\") can be used as a single-octet + quoting mechanism within quoted-string and comment constructs. + Recipients that process the value of a quoted-string MUST handle a + quoted-pair as if it were replaced by the octet following the backslash.¶

+
+ +
+
  quoted-pair    = "\" ( HTAB / SP / VCHAR / obs-text )
+
¶ +
+

+ A sender SHOULD NOT generate a quoted-pair in a quoted-string except + where necessary to quote DQUOTE and backslash octets occurring within that + string. + A sender SHOULD NOT generate a quoted-pair in a comment except + where necessary to quote parentheses ["(" and ")"] and backslash octets + occurring within that comment.¶

+
+
+
+
+

+5.6.5. Comments +

+
+

+ + + Comments can be included in some HTTP fields by surrounding + the comment text with parentheses. Comments are only allowed in + fields containing "comment" as part of their field value definition.¶

+
+ + +
+
  comment        = "(" *( ctext / quoted-pair / comment ) ")"
+  ctext          = HTAB / SP / %x21-27 / %x2A-5B / %x5D-7E / obs-text
+
¶ +
+
+
+
+
+

+5.6.6. Parameters +

+
+

+ + + + Parameters are instances of name/value pairs; they are often used in field + values as a common syntax for appending auxiliary information to an item. + Each parameter is usually delimited by an immediately preceding semicolon.¶

+
+ + + + +
+
  parameters      = *( OWS ";" OWS [ parameter ] )
+  parameter       = parameter-name "=" parameter-value
+  parameter-name  = token
+  parameter-value = ( token / quoted-string )
+
¶ +
+

+ Parameter names are case-insensitive. Parameter values might or might + not be case-sensitive, depending on the semantics of the parameter + name. Examples of parameters and some equivalent forms can be seen in + media types (Section 8.3.1) and the Accept header field + (Section 12.5.1).¶

+

+ A parameter value that matches the token production can be + transmitted either as a token or within a quoted-string. The quoted and + unquoted values are equivalent.¶

+ +
+
+
+
+

+5.6.7. Date/Time Formats +

+ +

+ Prior to 1995, there were three different formats commonly used by servers + to communicate timestamps. For compatibility with old implementations, all + three are defined here. The preferred format is a fixed-length and + single-zone subset of the date and time specification used by the + Internet Message Format [RFC5322].¶

+ +
+
  HTTP-date    = IMF-fixdate / obs-date
+
¶ +
+

+ An example of the preferred format is¶

+
+
+  Sun, 06 Nov 1994 08:49:37 GMT    ; IMF-fixdate
+
¶ +
+

+ Examples of the two obsolete formats are¶

+
+
+  Sunday, 06-Nov-94 08:49:37 GMT   ; obsolete RFC 850 format
+  Sun Nov  6 08:49:37 1994         ; ANSI C's asctime() format
+
¶ +
+

+ A recipient that parses a timestamp value in an HTTP field MUST + accept all three HTTP-date formats. When a sender generates a field + that contains one or more timestamps defined as HTTP-date, + the sender MUST generate those timestamps in the IMF-fixdate format.¶

+

+ An HTTP-date value represents time as an instance of Coordinated + Universal Time (UTC). The first two formats indicate UTC by the + three-letter abbreviation for Greenwich Mean Time, "GMT", a predecessor + of the UTC name; values in the asctime format are assumed to be in UTC.¶

+

+ A "clock" is an implementation capable of providing a + reasonable approximation of the current instant in UTC. + A clock implementation ought to use NTP ([RFC5905]), + or some similar protocol, to synchronize with UTC.¶

+
+

+ + + + + + + + + + + Preferred format:¶

+
+ + + + + + + + + + + + +
+
  IMF-fixdate  = day-name "," SP date1 SP time-of-day SP GMT
+  ; fixed length/zone/capitalization subset of the format
+  ; see Section 3.3 of [RFC5322]
+
+  day-name     = %s"Mon" / %s"Tue" / %s"Wed"
+               / %s"Thu" / %s"Fri" / %s"Sat" / %s"Sun"
+
+  date1        = day SP month SP year
+               ; e.g., 02 Jun 1982
+
+  day          = 2DIGIT
+  month        = %s"Jan" / %s"Feb" / %s"Mar" / %s"Apr"
+               / %s"May" / %s"Jun" / %s"Jul" / %s"Aug"
+               / %s"Sep" / %s"Oct" / %s"Nov" / %s"Dec"
+  year         = 4DIGIT
+
+  GMT          = %s"GMT"
+
+  time-of-day  = hour ":" minute ":" second
+               ; 00:00:00 - 23:59:60 (leap second)
+
+  hour         = 2DIGIT
+  minute       = 2DIGIT
+  second       = 2DIGIT
+
¶ +
+
+

+ + + + + + + + Obsolete formats:¶

+
+ +
+
  obs-date     = rfc850-date / asctime-date
+
¶ +
+ +
+
  rfc850-date  = day-name-l "," SP date2 SP time-of-day SP GMT
+  date2        = day "-" month "-" 2DIGIT
+               ; e.g., 02-Jun-82
+
+  day-name-l   = %s"Monday" / %s"Tuesday" / %s"Wednesday"
+               / %s"Thursday" / %s"Friday" / %s"Saturday"
+               / %s"Sunday"
+
¶ +
+ +
+
  asctime-date = day-name SP date3 SP time-of-day SP year
+  date3        = month SP ( 2DIGIT / ( SP 1DIGIT ))
+               ; e.g., Jun  2
+
¶ +
+

+ HTTP-date is case sensitive. Note that Section 4.2 of [CACHING] relaxes this for cache recipients.¶

+

+ A sender MUST NOT generate additional whitespace in an HTTP-date beyond + that specifically included as SP in the grammar. + The semantics of day-name, day, + month, year, and time-of-day + are the same as those defined for the Internet Message Format constructs + with the corresponding name ([RFC5322], Section 3.3).¶

+

+ Recipients of a timestamp value in rfc850-date format, which uses a + two-digit year, MUST interpret a timestamp that appears to be more + than 50 years in the future as representing the most recent year in the + past that had the same last two digits.¶

+

+ Recipients of timestamp values are encouraged to be robust in parsing + timestamps unless otherwise restricted by the field definition. + For example, messages are occasionally forwarded over HTTP from a non-HTTP + source that might generate any of the date and time specifications defined + by the Internet Message Format.¶

+ +
+
+
+
+
+
+
+
+

+6. Message Abstraction +

+ + + +

+ Each major version of HTTP defines its own syntax for communicating + messages. This section defines an abstract data type for HTTP messages + based on a generalization of those message characteristics, common structure, + and capacity for conveying semantics. This abstraction is used to define + requirements on senders and recipients that are independent of the HTTP + version, such that a message in one version can be relayed through other + versions without changing its meaning.¶

+

+ A "message" consists of the following:¶

+
    +
  • control data to describe and route the message,¶ +
    • +
  • a headers lookup table of name/value pairs for extending that control + data and conveying additional information about the sender, message, + content, or context,¶ +
    • +
  • a potentially unbounded stream of content, and¶ +
    • +
  • a trailers lookup table of name/value pairs for communicating information + obtained while sending the content.¶ +
    • +
+

+ Framing and control data is sent first, followed by a header section + containing fields for the headers table. When a message includes content, + the content is sent after the header section, potentially followed by a + trailer section that might contain fields for the trailers table.¶

+

+ Messages are expected to be processed as a stream, wherein the purpose of + that stream and its continued processing is revealed while being read. + Hence, control data describes what the recipient needs to know immediately, + header fields describe what needs to be known before receiving content, + the content (when present) presumably contains what the recipient wants or + needs to fulfill the message semantics, and trailer fields provide + optional metadata that was unknown prior to sending the content.¶

+

+ Messages are intended to be "self-descriptive": + everything a recipient needs to know about the message can be determined by + looking at the message itself, after decoding or reconstituting parts that + have been compressed or elided in transit, without requiring an + understanding of the sender's current application state (established via + prior messages). However, a client MUST retain knowledge of the request when + parsing, interpreting, or caching a corresponding response. For example, + responses to the HEAD method look just like the beginning of a + response to GET but cannot be parsed in the same manner.¶

+

+ Note that this message abstraction is a generalization across many versions + of HTTP, including features that might not be found in some versions. For + example, trailers were introduced within the HTTP/1.1 chunked transfer + coding as a trailer section after the content. An equivalent feature is + present in HTTP/2 and HTTP/3 within the header block that terminates each + stream.¶

+
+
+

+6.1. Framing and Completeness +

+ + +

+ Message framing indicates how each message begins and ends, such that each + message can be distinguished from other messages or noise on the same + connection. Each major version of HTTP defines its own framing mechanism.¶

+

+ HTTP/0.9 and early deployments of HTTP/1.0 used closure of the underlying + connection to end a response. For backwards compatibility, this implicit + framing is also allowed in HTTP/1.1. However, implicit framing can fail to + distinguish an incomplete response if the connection closes early. For + that reason, almost all modern implementations use explicit framing in + the form of length-delimited sequences of message data.¶

+

+ A message is considered "complete" when all of the octets + indicated by its framing are available. Note that, + when no explicit framing is used, a response message that is ended + by the underlying connection's close is considered complete even though it + might be indistinguishable from an incomplete response, unless a + transport-level error indicates that it is not complete.¶

+
+
+
+
+

+6.2. Control Data +

+ +

+ Messages start with control data that describe its primary purpose. Request + message control data includes a request method (Section 9), + request target (Section 7.1), and protocol version + (Section 2.5). Response message control data includes + a status code (Section 15), optional reason phrase, and + protocol version.¶

+

+ In HTTP/1.1 ([HTTP/1.1]) and earlier, control data is sent + as the first line of a message. In HTTP/2 ([HTTP/2]) and + HTTP/3 ([HTTP/3]), control data is sent as pseudo-header + fields with a reserved name prefix (e.g., ":authority").¶

+

+ Every HTTP message has a protocol version. Depending on the version in use, + it might be identified within the message explicitly or inferred by the + connection over which the message is received. Recipients use that version + information to determine limitations or potential for later communication + with that sender.¶

+

+ When a message is forwarded by an intermediary, the protocol version is + updated to reflect the version used by that intermediary. + The Via header field (Section 7.6.3) is used to + communicate upstream protocol information within a forwarded message.¶

+

+ A client SHOULD send a request version equal to the highest + version to which the client is conformant and + whose major version is no higher than the highest version supported + by the server, if this is known. A client MUST NOT send a + version to which it is not conformant.¶

+

+ A client MAY send a lower request version if it is known that + the server incorrectly implements the HTTP specification, but only + after the client has attempted at least one normal request and determined + from the response status code or header fields (e.g., Server) that + the server improperly handles higher request versions.¶

+

+ A server SHOULD send a response version equal to the highest version to + which the server is conformant that has a major version less than or equal + to the one received in the request. + A server MUST NOT send a version to which it is not conformant. + A server can send a 505 (HTTP Version Not Supported) + response if it wishes, for any reason, to refuse service of the client's + major protocol version.¶

+

+ A recipient that receives a message with a major version number that it + implements and a minor version number higher than what it implements + SHOULD process the message as if it + were in the highest minor version within that major version to which the + recipient is conformant. A recipient can assume that a message with a + higher minor version, when sent to a recipient that has not yet indicated + support for that higher version, is sufficiently backwards-compatible to be + safely processed by any implementation of the same major version.¶

+
+
+
+
+

+6.3. Header Fields +

+ + +

+ Fields (Section 5) that are sent or received before the content + are referred to as "header fields" (or just "headers", colloquially).¶

+

+ The "header section" of a message consists of a sequence of + header field lines. Each header field might modify or extend message + semantics, describe the sender, define the content, or provide additional + context.¶

+ +
+
+
+
+

+6.4. Content +

+ +

+ HTTP messages often transfer a complete or partial representation as the + message "content": a stream of octets sent after the header + section, as delineated by the message framing.¶

+

+ This abstract definition of content reflects the data after it has been + extracted from the message framing. For example, an HTTP/1.1 message body + (Section 6 of [HTTP/1.1]) might consist of a stream of data encoded + with the chunked transfer coding -- a sequence of data chunks, one + zero-length chunk, and a trailer section -- whereas + the content of that same message + includes only the data stream after the transfer coding has been decoded; + it does not include the chunk lengths, chunked framing syntax, nor the + trailer fields (Section 6.5).¶

+ +
+
+

+6.4.1. Content Semantics +

+

+ The purpose of content in a request is defined by the method semantics + (Section 9).¶

+

+ For example, a representation in the content of a PUT request + (Section 9.3.4) represents the desired state of the + target resource after the request is successfully applied, + whereas a representation in the content of a POST request + (Section 9.3.3) represents information to be processed by the + target resource.¶

+

+ In a response, the content's purpose is defined by the request method, + response status code (Section 15), and response + fields describing that content. + For example, the content of a 200 (OK) response to GET + (Section 9.3.1) represents the current state of the + target resource, as observed at the time of the message + origination date (Section 6.6.1), whereas the content of + the same status code in a response to POST might represent either the + processing result or the new state of the target resource after applying + the processing.¶

+

+ The content of a 206 (Partial Content) response to GET + contains either a single part of the selected representation or a + multipart message body containing multiple parts of that representation, + as described in Section 15.3.7.¶

+

+ Response messages with an error status code usually contain content that + represents the error condition, such that the content describes the + error state and what steps are suggested for resolving it.¶

+

+ Responses to the HEAD request method (Section 9.3.2) never include + content; the associated response header fields indicate only + what their values would have been if the request method had been GET + (Section 9.3.1).¶

+

+ 2xx (Successful) responses to a CONNECT request method + (Section 9.3.6) switch the connection to tunnel mode instead of + having content.¶

+

+ All 1xx (Informational), 204 (No Content), and + 304 (Not Modified) responses do not include content.¶

+

+ All other responses do include content, although that content + might be of zero length.¶

+
+
+
+
+

+6.4.2. Identifying Content +

+

+ When a complete or partial representation is transferred as message + content, it is often desirable for the sender to supply, or the recipient + to determine, an identifier for a resource corresponding to that specific + representation. For example, a client making a GET request on a resource + for "the current weather report" might want an identifier specific to the + content returned (e.g., "weather report for Laguna Beach at 20210720T1711"). + This can be useful for sharing or bookmarking content from resources that + are expected to have changing representations over time.¶

+

+ For a request message:¶

+
    +
  • If the request has a Content-Location header field, + then the sender asserts that the content is a representation of the + resource identified by the Content-Location field value. However, + such an assertion cannot be trusted unless it can be verified by + other means (not defined by this specification). The information + might still be useful for revision history links.¶ +
    • +
  • Otherwise, the content is unidentified by HTTP, but a more specific + identifier might be supplied within the content itself.¶ +
    • +
+

+ For a response message, the following rules are applied in order until a + match is found:¶

+
    +
  1. If the request method is HEAD or the response status code is + 204 (No Content) or 304 (Not Modified), + there is no content in the response.¶ +
    2. +
  2. If the request method is GET and the response status code is + 200 (OK), + the content is a representation of the target resource (Section 7.1).¶ +
    3. +
  3. If the request method is GET and the response status code is + 203 (Non-Authoritative Information), the content is + a potentially modified or enhanced representation of the + target resource as provided by an intermediary.¶ +
    4. +
  4. If the request method is GET and the response status code is + 206 (Partial Content), + the content is one or more parts of a representation of the + target resource.¶ +
    5. +
  5. If the response has a Content-Location header field + and its field value is a reference to the same URI as the target URI, + the content is a representation of the target resource.¶ +
    6. +
  6. If the response has a Content-Location header field + and its field value is a reference to a URI different from the + target URI, then the sender asserts that the content is a + representation of the resource identified by the Content-Location + field value. However, such an assertion cannot be trusted unless + it can be verified by other means (not defined by this specification).¶ +
    7. +
  7. Otherwise, the content is unidentified by HTTP, but a more specific + identifier might be supplied within the content itself.¶ +
    8. +
+
+
+
+
+
+
+

+6.5. Trailer Fields +

+ + + +

+ Fields (Section 5) that are located within a + "trailer section" are referred to as "trailer fields" + (or just "trailers", colloquially). + Trailer fields can be useful for supplying message integrity checks, digital + signatures, delivery metrics, or post-processing status information.¶

+

+ Trailer fields ought to be processed and stored separately from the fields + in the header section to avoid contradicting message semantics known at + the time the header section was complete. The presence or absence of + certain header fields might impact choices made for the routing or + processing of the message as a whole before the trailers are received; + those choices cannot be unmade by the later discovery of trailer fields.¶

+
+
+

+6.5.1. Limitations on Use of Trailers +

+

+ A trailer section is only possible when supported by the version + of HTTP in use and enabled by an explicit framing mechanism. + For example, the chunked transfer coding in HTTP/1.1 allows a trailer section to be + sent after the content (Section 7.1.2 of [HTTP/1.1]).¶

+

+ Many fields cannot be processed outside the header section because + their evaluation is necessary prior to receiving the content, such as + those that describe message framing, routing, authentication, + request modifiers, response controls, or content format. + A sender MUST NOT generate a trailer field unless the sender knows the + corresponding header field name's definition permits the field to be sent + in trailers.¶

+

+ Trailer fields can be difficult to process by intermediaries that forward + messages from one protocol version to another. If the entire message can be + buffered in transit, some intermediaries could merge trailer fields into + the header section (as appropriate) before it is forwarded. However, in + most cases, the trailers are simply discarded. + A recipient MUST NOT merge a trailer field into a header section unless + the recipient understands the corresponding header field definition and + that definition explicitly permits and defines how trailer field values + can be safely merged.¶

+

+ The presence of the keyword "trailers" in the TE header field (Section 10.1.4) of a request indicates that the client is willing to + accept trailer fields, on behalf of itself and any downstream clients. For + requests from an intermediary, this implies that all + downstream clients are willing to accept trailer fields in the forwarded + response. Note that the presence of "trailers" does not mean that the + client(s) will process any particular trailer field in the response; only + that the trailer section(s) will not be dropped by any of the clients.¶

+

+ Because of the potential for trailer fields to be discarded in transit, a + server SHOULD NOT generate trailer fields that it believes are necessary + for the user agent to receive.¶

+
+
+
+
+

+6.5.2. Processing Trailer Fields +

+

+ The "Trailer" header field (Section 6.6.2) can be sent + to indicate fields likely to be sent in the trailer section, which allows + recipients to prepare for their receipt before processing the content. + For example, this could be useful if a field name indicates that a dynamic + checksum should be calculated as the content is received and then + immediately checked upon receipt of the trailer field value.¶

+

+ Like header fields, trailer fields with the same name are processed in the + order received; multiple trailer field lines with the same name have the + equivalent semantics as appending the multiple values as a list of members. + Trailer fields that might be generated more than once during a message + MUST be defined as a list-based field even if each member value is only + processed once per field line received.¶

+

+ At the end of a message, a recipient MAY treat the set of received + trailer fields as a data structure of name/value pairs, similar to (but + separate from) the header fields. Additional processing expectations, if + any, can be defined within the field specification for a field intended + for use in trailers.¶

+
+
+
+
+
+
+

+6.6. Message Metadata +

+

+ Fields that describe the message itself, such as when and how the + message has been generated, can appear in both requests and responses.¶

+
+
+

+6.6.1. Date +

+ + + +

+ The "Date" header field represents the date and time at which + the message was originated, having the same semantics as the Origination + Date Field (orig-date) defined in Section 3.6.1 of [RFC5322]. + The field value is an HTTP-date, as defined in Section 5.6.7.¶

+ +
+
  Date = HTTP-date
+
¶ +
+

+ An example is¶

+
+
Date: Tue, 15 Nov 1994 08:12:31 GMT
+
¶ +
+

+ A sender that generates a Date header field SHOULD generate its + field value as the best available approximation of the date and time of + message generation. In theory, the date ought to represent the moment just + before generating the message content. In practice, a sender can generate + the date value at any time during message origination.¶

+

+ An origin server with a clock (as defined in + Section 5.6.7) MUST generate a Date header field in + all 2xx (Successful), 3xx (Redirection), + and 4xx (Client Error) responses, + and MAY generate a Date header field in + 1xx (Informational) and + 5xx (Server Error) responses.¶

+

+ An origin server without a clock MUST NOT generate a Date header field.¶

+

+ A recipient with a clock that receives a response message without a Date + header field MUST record the time it was received and append a + corresponding Date header field to the message's header section if it is + cached or forwarded downstream.¶

+

+ A recipient with a clock that receives a response with an invalid Date + header field value MAY replace that value with the time that + response was received.¶

+

+ A user agent MAY send a Date header field in a request, though generally + will not do so unless it is believed to convey useful information to the + server. For example, custom applications of HTTP might convey a Date if + the server is expected to adjust its interpretation of the user's request + based on differences between the user agent and server clocks.¶

+
+
+
+
+

+6.6.2. Trailer +

+ + + +

+ The "Trailer" header field provides a list of field names that the sender + anticipates sending as trailer fields within that message. This allows a + recipient to prepare for receipt of the indicated metadata before it starts + processing the content.¶

+ + +
+
  Trailer = #field-name
+
¶ +
+

+ For example, a sender might indicate that a signature will + be computed as the content is being streamed and provide the final + signature as a trailer field. This allows a recipient to perform the same + check on the fly as it receives the content.¶

+

+ A sender that intends to generate one or more trailer fields in a message + SHOULD generate a Trailer header field in the header + section of that message to indicate which fields might be present in the + trailers.¶

+

+ If an intermediary discards the trailer section in transit, the + Trailer field could provide a hint of what metadata + was lost, though there is no guarantee that a sender of Trailer + will always follow through by sending the named fields.¶

+
+
+
+
+
+
+
+
+

+7. Routing HTTP Messages +

+

+ HTTP request message routing is determined by each client based on the + target resource, the client's proxy configuration, and + establishment or reuse of an inbound connection. The corresponding + response routing follows the same connection chain back to the client.¶

+
+
+

+7.1. Determining the Target Resource +

+ + + +

+ Although HTTP is used in a wide variety of applications, most clients rely + on the same resource identification mechanism and configuration techniques + as general-purpose Web browsers. Even when communication options are + hard-coded in a client's configuration, we can think of their combined + effect as a URI reference (Section 4.1).¶

+

+ A URI reference is resolved to its absolute form in order to obtain the + "target URI". The target URI excludes the reference's + fragment component, if any, since fragment identifiers are reserved for + client-side processing ([URI], Section 3.5).¶

+

+ To perform an action on a "target resource", the client sends + a request message containing enough components of its parsed target URI to + enable recipients to identify that same resource. For historical reasons, + the parsed target URI components, collectively referred to as the + "request target", are sent within the message control data + and the Host header field (Section 7.2).¶

+

+ There are two unusual cases for which the request target components are in + a method-specific form:¶

+
    +
  • + For CONNECT (Section 9.3.6), the request target is the host + name and port number of the tunnel destination, separated by a colon.¶ +
    • +
  • + For OPTIONS (Section 9.3.7), the request target can be a + single asterisk ("*").¶ +
    • +
+

+ See the respective method definitions for details. These forms MUST NOT + be used with other methods.¶

+

+ Upon receipt of a client's request, a server reconstructs the target URI + from the received components in accordance with their local configuration + and incoming connection context. This reconstruction is specific to each + major protocol version. For example, + Section 3.3 of [HTTP/1.1] defines how a server + determines the target URI of an HTTP/1.1 request.¶

+
+ +
+
+
+
+
+

+7.2. Host and :authority +

+ + + +

+ The "Host" header field in a request provides the host and port + information from the target URI, enabling the origin + server to distinguish among resources while servicing requests + for multiple host names.¶

+

+ In HTTP/2 [HTTP/2] and HTTP/3 [HTTP/3], the + Host header field is, in some cases, supplanted by the ":authority" + pseudo-header field of a request's control data.¶

+ +
+
  Host = uri-host [ ":" port ] ; Section 4
+
¶ +
+

+ The target URI's authority information is critical for handling a + request. A user agent MUST generate a Host header field in a request + unless it sends that information as an ":authority" pseudo-header field. + A user agent that sends Host SHOULD send it as the first field in the + header section of a request.¶

+

+ For example, a GET request to the origin server for + <http://www.example.org/pub/WWW/> would begin with:¶

+
+
GET /pub/WWW/ HTTP/1.1
+Host: www.example.org
+
¶ +
+

+ Since the host and port information acts as an application-level routing + mechanism, it is a frequent target for malware seeking to poison + a shared cache or redirect a request to an unintended server. + An interception proxy is particularly vulnerable if it relies on + the host and port information for redirecting requests to internal + servers, or for use as a cache key in a shared cache, without + first verifying that the intercepted connection is targeting a + valid IP address for that host.¶

+
+
+
+
+

+7.3. Routing Inbound Requests +

+

+ Once the target URI and its origin are determined, a client decides whether + a network request is necessary to accomplish the desired semantics and, + if so, where that request is to be directed.¶

+
+
+

+7.3.1. To a Cache +

+

+ If the client has a cache [CACHING] and the request can be + satisfied by it, then the request is + usually directed there first.¶

+
+
+
+
+

+7.3.2. To a Proxy +

+

+ If the request is not satisfied by a cache, then a typical client will + check its configuration to determine whether a proxy is to be used to + satisfy the request. Proxy configuration is implementation-dependent, + but is often based on URI prefix matching, selective authority matching, + or both, and the proxy itself is usually identified by an "http" or + "https" URI.¶

+

+ If an "http" or "https" proxy is applicable, the client connects + inbound by establishing (or reusing) a connection to that proxy and + then sending it an HTTP request message containing a request target + that matches the client's target URI.¶

+
+
+
+
+

+7.3.3. To the Origin +

+

+ If no proxy is applicable, a typical client will invoke a handler + routine (specific to the target URI's scheme) to obtain access to the + identified resource. How that is accomplished is dependent on the + target URI scheme and defined by its associated specification.¶

+

+ Section 4.3.2 defines how to obtain access to an + "http" resource by establishing (or reusing) an inbound connection to + the identified origin server and then sending it an HTTP request message + containing a request target that matches the client's target URI.¶

+

+ Section 4.3.3 defines how to obtain access to an + "https" resource by establishing (or reusing) an inbound secured + connection to an origin server that is authoritative for the identified + origin and then sending it an HTTP request message containing a request + target that matches the client's target URI.¶

+
+
+
+
+
+
+

+7.4. Rejecting Misdirected Requests +

+

+ Once a request is received by a server and parsed sufficiently to determine + its target URI, the server decides whether to process the request itself, + forward the request to another server, redirect the client to a different + resource, respond with an error, or drop the connection. This decision can + be influenced by anything about the request or connection context, but is + specifically directed at whether the server has been configured to process + requests for that target URI and whether the connection context is + appropriate for that request.¶

+

+ For example, a request might have been misdirected, + deliberately or accidentally, such that the information within a received + Host header field differs from the connection's host or port. + If the connection is from a trusted gateway, such inconsistency might + be expected; otherwise, it might indicate an attempt to bypass security + filters, trick the server into delivering non-public content, or poison a + cache. See Section 17 for security + considerations regarding message routing.¶

+

+ Unless the connection is from a trusted gateway, + an origin server MUST reject a request if any scheme-specific requirements + for the target URI are not met. In particular, + a request for an "https" resource MUST be rejected unless it has been + received over a connection that has been secured via a certificate + valid for that target URI's origin, as defined by Section 4.2.2.¶

+

+ The 421 (Misdirected Request) status code in a response + indicates that the origin server has rejected the request because it + appears to have been misdirected (Section 15.5.20).¶

+
+
+
+
+

+7.5. Response Correlation +

+

+ A connection might be used for multiple request/response exchanges. The + mechanism used to correlate between request and response messages is + version dependent; some versions of HTTP use implicit ordering of + messages, while others use an explicit identifier.¶

+

+ All responses, regardless of the status code (including interim + responses) can be sent at any time after a request is received, even if the + request is not yet complete. A response can complete before its + corresponding request is complete (Section 6.1). Likewise, clients are not expected + to wait any specific amount of time for a response. Clients + (including intermediaries) might abandon a request if the response is not + received within a reasonable period of time.¶

+

+ A client that receives a response while it is still sending the associated + request SHOULD continue sending that request unless it receives + an explicit indication to the contrary (see, e.g., Section 9.5 of [HTTP/1.1] and Section 6.4 of [HTTP/2]).¶

+
+
+
+
+

+7.6. Message Forwarding +

+

+ As described in Section 3.7, intermediaries can serve + a variety of roles in the processing of HTTP requests and responses. + Some intermediaries are used to improve performance or availability. + Others are used for access control or to filter content. + Since an HTTP stream has characteristics similar to a pipe-and-filter + architecture, there are no inherent limits to the extent an intermediary + can enhance (or interfere) with either direction of the stream.¶

+

+ Intermediaries are expected to forward messages even when protocol elements + are not recognized (e.g., new methods, status codes, or field names) since that + preserves extensibility for downstream recipients.¶

+

+ An intermediary not acting as a tunnel MUST implement the + Connection header field, as specified in + Section 7.6.1, and exclude fields from being forwarded + that are only intended for the incoming connection.¶

+

+ An intermediary MUST NOT forward a message to itself unless it is + protected from an infinite request loop. In general, an intermediary ought + to recognize its own server names, including any aliases, local variations, + or literal IP addresses, and respond to such requests directly.¶

+

+ An HTTP message can be parsed as a stream for incremental processing or + forwarding downstream. + However, senders and recipients cannot rely on incremental + delivery of partial messages, since some implementations will buffer or + delay message forwarding for the sake of network efficiency, security + checks, or content transformations.¶

+
+
+

+7.6.1. Connection +

+ + + + + +

+ The "Connection" header field allows the sender to list desired + control options for the current connection.¶

+
+
  Connection        = #connection-option
+  connection-option = token
+
¶ +
+

+ Connection options are case-insensitive.¶

+

+ When a field aside from Connection is used to supply control + information for or about the current connection, the sender MUST list + the corresponding field name within the Connection header field. + Note that some versions of HTTP prohibit the use of fields for such + information, and therefore do not allow the Connection field.¶

+

+ Intermediaries MUST parse a received Connection + header field before a message is forwarded and, for each + connection-option in this field, remove any header or trailer field(s) from + the message with the same name as the connection-option, and then + remove the Connection header field itself (or replace it with the + intermediary's own control options for the forwarded message).¶

+

+ Hence, the Connection header field provides a declarative way of + distinguishing fields that are only intended for the + immediate recipient ("hop-by-hop") from those fields that are + intended for all recipients on the chain ("end-to-end"), enabling the + message to be self-descriptive and allowing future connection-specific + extensions to be deployed without fear that they will be blindly + forwarded by older intermediaries.¶

+

+ Furthermore, intermediaries SHOULD remove or replace fields + that are known to require removal before forwarding, whether or not they appear as a + connection-option, after applying those fields' semantics. This includes but is not limited to:¶

+ +

+ A sender MUST NOT send a connection option corresponding to a + field that is intended for all recipients of the content. + For example, Cache-Control is never appropriate as a + connection option (Section 5.2 of [CACHING]).¶

+

+ Connection options do not always correspond to a field + present in the message, since a connection-specific field + might not be needed if there are no parameters associated with a + connection option. In contrast, a connection-specific field + received without a corresponding connection option usually indicates + that the field has been improperly forwarded by an intermediary and + ought to be ignored by the recipient.¶

+

+ When defining a new connection option that does not correspond to a field, + specification authors ought to reserve the corresponding field name + anyway in order to avoid later collisions. Such reserved field names are + registered in the "Hypertext Transfer Protocol (HTTP) Field Name Registry" + (Section 16.3.1).¶

+
+
+
+
+

+7.6.2. Max-Forwards +

+ + + +

+ The "Max-Forwards" header field provides a mechanism with the + TRACE (Section 9.3.8) and OPTIONS (Section 9.3.7) + request methods to limit the number of times that the request is forwarded by + proxies. This can be useful when the client is attempting to + trace a request that appears to be failing or looping mid-chain.¶

+ +
+
  Max-Forwards = 1*DIGIT
+
¶ +
+

+ The Max-Forwards value is a decimal integer indicating the remaining + number of times this request message can be forwarded.¶

+

+ Each intermediary that receives a TRACE or OPTIONS request containing a + Max-Forwards header field MUST check and update its value prior to + forwarding the request. If the received value is zero (0), the intermediary + MUST NOT forward the request; instead, the intermediary MUST respond as + the final recipient. If the received Max-Forwards value is greater than + zero, the intermediary MUST generate an updated Max-Forwards field in the + forwarded message with a field value that is the lesser of a) the received + value decremented by one (1) or b) the recipient's maximum supported value + for Max-Forwards.¶

+

+ A recipient MAY ignore a Max-Forwards header field received with any + other request methods.¶

+
+
+
+
+

+7.6.3. Via +

+ + + +

+ The "Via" header field indicates the presence of intermediate protocols and + recipients between the user agent and the server (on requests) or between + the origin server and the client (on responses), similar to the + "Received" header field in email + (Section 3.6.7 of [RFC5322]). + Via can be used for tracking message forwards, + avoiding request loops, and identifying the protocol capabilities of + senders along the request/response chain.¶

+ + + + + + +
+
  Via = #( received-protocol RWS received-by [ RWS comment ] )
+
+  received-protocol = [ protocol-name "/" ] protocol-version
+                    ; see Section 7.8
+  received-by       = pseudonym [ ":" port ]
+  pseudonym         = token
+
¶ +
+

+ Each member of the Via field value represents a proxy or gateway that has + forwarded the message. Each intermediary appends its own information + about how the message was received, such that the end result is ordered + according to the sequence of forwarding recipients.¶

+

+ A proxy MUST send an appropriate Via header field, as described below, in + each message that it forwards. + An HTTP-to-HTTP gateway MUST send an appropriate Via header field in + each inbound request message and MAY send a Via header field in + forwarded response messages.¶

+

+ For each intermediary, the received-protocol indicates the protocol and + protocol version used by the upstream sender of the message. Hence, the + Via field value records the advertised protocol capabilities of the + request/response chain such that they remain visible to downstream + recipients; this can be useful for determining what backwards-incompatible + features might be safe to use in response, or within a later request, as + described in Section 2.5. For brevity, the protocol-name + is omitted when the received protocol is HTTP.¶

+

+ The received-by portion is normally the host and optional + port number of a recipient server or client that subsequently forwarded the + message. + However, if the real host is considered to be sensitive information, a + sender MAY replace it with a pseudonym. If a port is not provided, + a recipient MAY interpret that as meaning it was received on the default + port, if any, for the received-protocol.¶

+

+ A sender MAY generate comments to identify the + software of each recipient, analogous to the User-Agent and + Server header fields. However, comments in Via + are optional, and a recipient MAY remove them prior to forwarding the + message.¶

+

+ For example, a request message could be sent from an HTTP/1.0 user + agent to an internal proxy code-named "fred", which uses HTTP/1.1 to + forward the request to a public proxy at p.example.net, which completes + the request by forwarding it to the origin server at www.example.com. + The request received by www.example.com would then have the following + Via header field:¶

+
+
Via: 1.0 fred, 1.1 p.example.net
+
¶ +
+

+ An intermediary used as a portal through a network firewall + SHOULD NOT forward the names and ports of hosts within the firewall + region unless it is explicitly enabled to do so. If not enabled, such an + intermediary SHOULD replace each received-by host of any host behind the + firewall by an appropriate pseudonym for that host.¶

+

+ An intermediary MAY combine an ordered subsequence of Via header + field list members into a single member if the entries have identical + received-protocol values. For example,¶

+
+
Via: 1.0 ricky, 1.1 ethel, 1.1 fred, 1.0 lucy
+
¶ +
+

+ could be collapsed to¶

+
+
Via: 1.0 ricky, 1.1 mertz, 1.0 lucy
+
¶ +
+

+ A sender SHOULD NOT combine multiple list members unless they are all + under the same organizational control and the hosts have already been + replaced by pseudonyms. A sender MUST NOT combine members that + have different received-protocol values.¶

+
+
+
+
+
+
+

+7.7. Message Transformations +

+ + +

+ Some intermediaries include features for transforming messages and their + content. A proxy might, for example, convert between image formats in + order to save cache space or to reduce the amount of traffic on a slow + link. However, operational problems might occur when these transformations + are applied to content intended for critical applications, such as medical + imaging or scientific data analysis, particularly when integrity checks or + digital signatures are used to ensure that the content received is + identical to the original.¶

+

+ An HTTP-to-HTTP proxy is called a "transforming proxy" + if it is designed or configured to modify messages in a semantically + meaningful way (i.e., modifications, beyond those required by normal + HTTP processing, that change the message in a way that would be + significant to the original sender or potentially significant to + downstream recipients). For example, a transforming proxy might be + acting as a shared annotation server (modifying responses to include + references to a local annotation database), a malware filter, a + format transcoder, or a privacy filter. Such transformations are presumed + to be desired by whichever client (or client organization) chose the + proxy.¶

+

+ If a proxy receives a target URI with a host name that is not a + fully qualified domain name, it MAY add its own domain to the host name + it received when forwarding the request. A proxy MUST NOT change the + host name if the target URI contains a fully qualified domain name.¶

+

+ A proxy MUST NOT modify the "absolute-path" and "query" parts of the + received target URI when forwarding it to the next inbound server except + as required by that forwarding protocol. For example, a proxy forwarding + a request to an origin server via HTTP/1.1 will replace an empty path with + "/" (Section 3.2.1 of [HTTP/1.1]) or "*" (Section 3.2.4 of [HTTP/1.1]), + depending on the request method.¶

+

+ A proxy MUST NOT transform the content (Section 6.4) of a + response message that contains a no-transform cache directive + (Section 5.2.2.6 of [CACHING]). Note that this + does not apply to message transformations that do not change the content, + such as the addition or removal of transfer codings + (Section 7 of [HTTP/1.1]).¶

+

+ A proxy MAY transform the content of a message + that does not contain a no-transform cache directive. + A proxy that transforms the content of a 200 (OK) response + can inform downstream recipients that a transformation has been + applied by changing the response status code to + 203 (Non-Authoritative Information) (Section 15.3.4).¶

+

+ A proxy SHOULD NOT modify header fields that provide information about + the endpoints of the communication chain, the resource state, or the + selected representation (other than the content) unless the field's + definition specifically allows such modification or the modification is + deemed necessary for privacy or security.¶

+
+
+
+
+

+7.8. Upgrade +

+ + + +

+ The "Upgrade" header field is intended to provide a simple mechanism + for transitioning from HTTP/1.1 to some other protocol on the same + connection.¶

+

+ A client MAY send a list of protocol names in the Upgrade header field + of a request to invite the server to switch to one or more of the named + protocols, in order of descending preference, before sending + the final response. A server MAY ignore a received Upgrade header field + if it wishes to continue using the current protocol on that connection. + Upgrade cannot be used to insist on a protocol change.¶

+ +
+
  Upgrade          = #protocol
+
+  protocol         = protocol-name ["/" protocol-version]
+  protocol-name    = token
+  protocol-version = token
+
¶ +
+

+ Although protocol names are registered with a preferred case, + recipients SHOULD use case-insensitive comparison when matching each + protocol-name to supported protocols.¶

+

+ A server that sends a 101 (Switching Protocols) response + MUST send an Upgrade header field to indicate the new protocol(s) to + which the connection is being switched; if multiple protocol layers are + being switched, the sender MUST list the protocols in layer-ascending + order. A server MUST NOT switch to a protocol that was not indicated by + the client in the corresponding request's Upgrade header field. + A server MAY choose to ignore the order of preference indicated by the + client and select the new protocol(s) based on other factors, such as the + nature of the request or the current load on the server.¶

+

+ A server that sends a 426 (Upgrade Required) response + MUST send an Upgrade header field to indicate the acceptable protocols, + in order of descending preference.¶

+

+ A server MAY send an Upgrade header field in any other response to + advertise that it implements support for upgrading to the listed protocols, + in order of descending preference, when appropriate for a future request.¶

+

+ The following is a hypothetical example sent by a client:¶

+
+
GET /hello HTTP/1.1
+Host: www.example.com
+Connection: upgrade
+Upgrade: websocket, IRC/6.9, RTA/x11
+
+
¶ +
+

+ The capabilities and nature of the + application-level communication after the protocol change is entirely + dependent upon the new protocol(s) chosen. However, immediately after + sending the 101 (Switching Protocols) response, the server is expected to continue responding to + the original request as if it had received its equivalent within the new + protocol (i.e., the server still has an outstanding request to satisfy + after the protocol has been changed, and is expected to do so without + requiring the request to be repeated).¶

+

+ For example, if the Upgrade header field is received in a GET request + and the server decides to switch protocols, it first responds + with a 101 (Switching Protocols) message in HTTP/1.1 and + then immediately follows that with the new protocol's equivalent of a + response to a GET on the target resource. This allows a connection to be + upgraded to protocols with the same semantics as HTTP without the + latency cost of an additional round trip. A server MUST NOT switch + protocols unless the received message semantics can be honored by the new + protocol; an OPTIONS request can be honored by any protocol.¶

+

+ The following is an example response to the above hypothetical request:¶

+
+
HTTP/1.1 101 Switching Protocols
+Connection: upgrade
+Upgrade: websocket
+
+[... data stream switches to websocket with an appropriate response
+(as defined by new protocol) to the "GET /hello" request ...]
+
¶ +
+

+ A sender of Upgrade MUST also send an "Upgrade" connection option in the + Connection header field (Section 7.6.1) + to inform intermediaries not to forward this field. + A server that receives an Upgrade header field in an HTTP/1.0 request + MUST ignore that Upgrade field.¶

+

+ A client cannot begin using an upgraded protocol on the connection until + it has completely sent the request message (i.e., the client can't change + the protocol it is sending in the middle of a message). + If a server receives both an Upgrade and an Expect header field + with the "100-continue" expectation (Section 10.1.1), the + server MUST send a 100 (Continue) response before sending + a 101 (Switching Protocols) response.¶

+

+ The Upgrade header field only applies to switching protocols on top of the + existing connection; it cannot be used to switch the underlying connection + (transport) protocol, nor to switch the existing communication to a + different connection. For those purposes, it is more appropriate to use a + 3xx (Redirection) response (Section 15.4).¶

+

+ This specification only defines the protocol name "HTTP" for use by + the family of Hypertext Transfer Protocols, as defined by the HTTP + version rules of Section 2.5 and future updates to this + specification. Additional protocol names ought to be registered using the + registration procedure defined in Section 16.7.¶

+
+
+
+
+
+
+

+8. Representation Data and Metadata +

+
+
+

+8.1. Representation Data +

+

+ The representation data associated with an HTTP message is + either provided as the content of the message or + referred to by the message semantics and the target + URI. The representation data is in a format and encoding defined by + the representation metadata header fields.¶

+

+ The data type of the representation data is determined via the header fields + Content-Type and Content-Encoding. + These define a two-layer, ordered encoding model:¶

+
+
+  representation-data := Content-Encoding( Content-Type( data ) )
+
¶ +
+
+
+
+
+

+8.2. Representation Metadata +

+

+ Representation header fields provide metadata about the representation. + When a message includes content, the representation header fields + describe how to interpret that data. In a response to a HEAD request, the + representation header fields describe the representation data that would + have been enclosed in the content if the same request had been a GET.¶

+
+
+
+
+

+8.3. Content-Type +

+ + + +

+ The "Content-Type" header field indicates the media type of the + associated representation: either the representation enclosed in + the message content or the selected representation, as determined by the + message semantics. The indicated media type defines both the data format + and how that data is intended to be processed by a recipient, within the + scope of the received message semantics, after any content codings + indicated by Content-Encoding are decoded.¶

+ +
+
  Content-Type = media-type
+
¶ +
+

+ Media types are defined in Section 8.3.1. An example of the + field is¶

+
+
Content-Type: text/html; charset=ISO-8859-4
+
¶ +
+

+ A sender that generates a message containing content SHOULD + generate a Content-Type header field in that message unless the intended + media type of the enclosed representation is unknown to the sender. + If a Content-Type header field is not present, the recipient MAY either + assume a media type of + "application/octet-stream" ([RFC2046], Section 4.5.1) + or examine the data to determine its type.¶

+

+ In practice, resource owners do not always properly configure their origin + server to provide the correct Content-Type for a given representation. + Some user agents examine the content and, in certain cases, + override the received type (for example, see [Sniffing]). + This "MIME sniffing" risks drawing incorrect conclusions about the data, + which might expose the user to additional security risks + (e.g., "privilege escalation"). + Furthermore, distinct media types often share a common data format, + differing only in how the data is intended to be processed, which is + impossible to distinguish by inspecting the data alone. + When sniffing is implemented, implementers are encouraged to provide a + means for the user to disable it.¶

+

+ Although Content-Type is defined as a singleton field, it is + sometimes incorrectly generated multiple times, resulting in a combined + field value that appears to be a list. + Recipients often attempt to handle this error by using the last + syntactically valid member of the list, leading to potential + interoperability and security issues if different implementations + have different error handling behaviors.¶

+
+
+

+8.3.1. Media Type +

+

+ HTTP uses media types [RFC2046] in the + Content-Type (Section 8.3) + and Accept (Section 12.5.1) header fields in + order to provide open and extensible data typing and type negotiation. + Media types define both a data format and various processing models: + how to process that data in accordance with the message context.¶

+ + + +
+
  media-type = type "/" subtype parameters
+  type       = token
+  subtype    = token
+
¶ +
+

+ The type and subtype tokens are case-insensitive.¶

+

+ The type/subtype MAY be followed by semicolon-delimited parameters + (Section 5.6.6) in the form of name/value pairs. + The presence or absence of a parameter might be significant to the + processing of a media type, depending on its definition within the media + type registry. + Parameter values might or might not be case-sensitive, depending on the + semantics of the parameter name.¶

+

+ For example, the following media types are equivalent in describing HTML + text data encoded in the UTF-8 character encoding scheme, but the first is + preferred for consistency (the "charset" parameter value is defined as + being case-insensitive in [RFC2046], Section 4.1.2):¶

+
+
+  text/html;charset=utf-8
+  Text/HTML;Charset="utf-8"
+  text/html; charset="utf-8"
+  text/html;charset=UTF-8
+
¶ +
+

+ Media types ought to be registered with IANA according to the + procedures defined in [BCP13].¶

+
+
+
+
+

+8.3.2. Charset +

+

+ HTTP uses "charset" names to indicate or negotiate the + character encoding scheme ([RFC6365], Section 2) + of a textual representation. In the fields defined by this document, + charset names appear either in parameters (Content-Type), + or, for Accept-Encoding, in the form of a plain token. + In both cases, charset names are matched case-insensitively.¶

+

+ Charset names ought to be registered in the IANA "Character Sets" registry + (<https://www.iana.org/assignments/character-sets>) + according to the procedures defined in Section 2 of [RFC2978].¶

+ +
+
+
+
+

+8.3.3. Multipart Types +

+

+ MIME provides for a number of "multipart" types -- encapsulations of + one or more representations within a single message body. All multipart + types share a common syntax, as defined in Section 5.1.1 of [RFC2046], + and include a boundary parameter as part of the media type + value. The message body is itself a protocol element; a sender MUST + generate only CRLF to represent line breaks between body parts.¶

+

+ HTTP message framing does not use the multipart boundary as an indicator + of message body length, though it might be used by implementations that + generate or process the content. For example, the "multipart/form-data" + type is often used for carrying form data in a request, as described in + [RFC7578], and the "multipart/byteranges" type is defined + by this specification for use in some 206 (Partial Content) + responses (see Section 15.3.7).¶

+
+
+
+
+
+
+

+8.4. Content-Encoding +

+ + + +

+ The "Content-Encoding" header field indicates what content codings + have been applied to the representation, beyond those inherent in the media + type, and thus what decoding mechanisms have to be applied in order to + obtain data in the media type referenced by the Content-Type + header field. + Content-Encoding is primarily used to allow a representation's data to be + compressed without losing the identity of its underlying media type.¶

+ +
+
  Content-Encoding = #content-coding
+
¶ +
+

+ An example of its use is¶

+
+
Content-Encoding: gzip
+
¶ +
+

+ If one or more encodings have been applied to a representation, the sender + that applied the encodings MUST generate a Content-Encoding header field + that lists the content codings in the order in which they were applied. + Note that the coding named "identity" is reserved for its special role + in Accept-Encoding and thus SHOULD NOT be included.¶

+

+ Additional information about the encoding parameters can be provided + by other header fields not defined by this specification.¶

+

+ Unlike Transfer-Encoding (Section 6.1 of [HTTP/1.1]), the codings listed + in Content-Encoding are a characteristic of the representation; the + representation is defined in terms of the coded form, and all other + metadata about the representation is about the coded form unless otherwise + noted in the metadata definition. Typically, the representation is only + decoded just prior to rendering or analogous usage.¶

+

+ If the media type includes an inherent encoding, such as a data format + that is always compressed, then that encoding would not be restated in + Content-Encoding even if it happens to be the same algorithm as one + of the content codings. Such a content coding would only be listed if, + for some bizarre reason, it is applied a second time to form the + representation. Likewise, an origin server might choose to publish the + same data as multiple representations that differ only in whether + the coding is defined as part of Content-Type or + Content-Encoding, since some user agents will behave differently in their + handling of each response (e.g., open a "Save as ..." dialog instead of + automatic decompression and rendering of content).¶

+

+ An origin server MAY respond with a status code of + 415 (Unsupported Media Type) if a representation in the + request message has a content coding that is not acceptable.¶

+
+
+

+8.4.1. Content Codings +

+ + + + + + +

+ Content coding values indicate an encoding transformation that has + been or can be applied to a representation. Content codings are primarily + used to allow a representation to be compressed or otherwise usefully + transformed without losing the identity of its underlying media type + and without loss of information. Frequently, the representation is stored + in coded form, transmitted directly, and only decoded by the final recipient.¶

+ +
+
  content-coding   = token
+
¶ +
+

+ All content codings are case-insensitive and ought to be registered + within the "HTTP Content Coding Registry", as described in + Section 16.6¶

+

+ Content-coding values are used in the + Accept-Encoding (Section 12.5.3) + and Content-Encoding (Section 8.4) + header fields.¶

+
+
+
+8.4.1.1. Compress Coding +
+ +

+ The "compress" coding is an adaptive Lempel-Ziv-Welch (LZW) coding + [Welch] that is commonly produced by the UNIX file + compression program "compress". + A recipient SHOULD consider "x-compress" to be equivalent to "compress".¶

+
+
+
+
+
+8.4.1.2. Deflate Coding +
+ +

+ The "deflate" coding is a "zlib" data format [RFC1950] + containing a "deflate" compressed data stream [RFC1951] + that uses a combination of the Lempel-Ziv (LZ77) compression algorithm and + Huffman coding.¶

+ +
+
+
+
+
+8.4.1.3. Gzip Coding +
+ +

+ The "gzip" coding is an LZ77 coding with a 32-bit Cyclic Redundancy Check + (CRC) that is commonly + produced by the gzip file compression program [RFC1952]. + A recipient SHOULD consider "x-gzip" to be equivalent to "gzip".¶

+
+
+
+
+
+
+
+
+

+8.5. Content-Language +

+ + + +

+ The "Content-Language" header field describes the natural + language(s) of the intended audience for the representation. Note that this might + not be equivalent to all the languages used within the representation.¶

+ +
+
  Content-Language = #language-tag
+
¶ +
+

+ Language tags are defined in Section 8.5.1. The primary purpose of + Content-Language is to allow a user to identify and differentiate + representations according to the users' own preferred language. Thus, if the + content is intended only for a Danish-literate audience, the + appropriate field is¶

+
+
Content-Language: da
+
¶ +
+

+ If no Content-Language is specified, the default is that the content + is intended for all language audiences. This might mean that the + sender does not consider it to be specific to any natural language, + or that the sender does not know for which language it is intended.¶

+

+ Multiple languages MAY be listed for content that is intended for + multiple audiences. For example, a rendition of the "Treaty of + Waitangi", presented simultaneously in the original Maori and English + versions, would call for¶

+
+
Content-Language: mi, en
+
¶ +
+

+ However, just because multiple languages are present within a representation + does not mean that it is intended for multiple linguistic audiences. + An example would be a beginner's language primer, such as "A First + Lesson in Latin", which is clearly intended to be used by an + English-literate audience. In this case, the Content-Language would + properly only include "en".¶

+

+ Content-Language MAY be applied to any media type -- it is not + limited to textual documents.¶

+
+
+

+8.5.1. Language Tags +

+

+ A language tag, as defined in [RFC5646], identifies a + natural language spoken, written, or otherwise conveyed by human beings for + communication of information to other human beings. Computer languages are + explicitly excluded.¶

+

+ HTTP uses language tags within the Accept-Language and + Content-Language header fields. + Accept-Language uses the broader language-range production + defined in Section 12.5.4, whereas + Content-Language uses the language-tag production defined + below.¶

+ +
+
  language-tag = <Language-Tag, see [RFC5646], Section 2.1>
+
¶ +
+

+ A language tag is a sequence of one or more case-insensitive subtags, each + separated by a hyphen character ("-", %x2D). In most cases, a language tag + consists of a primary language subtag that identifies a broad family of + related languages (e.g., "en" = English), which is optionally followed by a + series of subtags that refine or narrow that language's range (e.g., + "en-CA" = the variety of English as communicated in Canada). + Whitespace is not allowed within a language tag. + Example tags include:¶

+
+
+  fr, en-US, es-419, az-Arab, x-pig-latin, man-Nkoo-GN
+
¶ +
+

+ See [RFC5646] for further information.¶

+
+
+
+
+
+
+

+8.6. Content-Length +

+ + + +

+ The "Content-Length" header field indicates the associated representation's + data length as a decimal non-negative integer number of octets. + When transferring a representation as content, Content-Length refers + specifically to the amount of data enclosed so that it can be used to + delimit framing (e.g., Section 6.2 of [HTTP/1.1]). + In other cases, Content-Length indicates the selected representation's + current length, which can be used by recipients to estimate transfer time + or to compare with previously stored representations.¶

+ +
+
  Content-Length = 1*DIGIT
+
¶ +
+

+ An example is¶

+
+
Content-Length: 3495
+
¶ +
+

+ A user agent SHOULD send Content-Length in a request when the method + defines a meaning for enclosed content and it is not sending + Transfer-Encoding. + For example, a user agent normally sends Content-Length in a POST request + even when the value is 0 (indicating empty content). + A user agent SHOULD NOT send a + Content-Length header field when the request message does not contain + content and the method semantics do not anticipate such data.¶

+

+ A server MAY send a Content-Length header field in a response to a HEAD + request (Section 9.3.2); a server MUST NOT send Content-Length in such a + response unless its field value equals the decimal number of octets that + would have been sent in the content of a response if the same + request had used the GET method.¶

+

+ A server MAY send a Content-Length header field in a + 304 (Not Modified) response to a conditional GET request + (Section 15.4.5); a server MUST NOT send Content-Length in such a + response unless its field value equals the decimal number of octets that + would have been sent in the content of a 200 (OK) + response to the same request.¶

+

+ A server MUST NOT send a Content-Length header field in any response + with a status code of + 1xx (Informational) or 204 (No Content). + A server MUST NOT send a Content-Length header field in any + 2xx (Successful) response to a CONNECT request (Section 9.3.6).¶

+

+ Aside from the cases defined above, in the absence of Transfer-Encoding, + an origin server SHOULD send a Content-Length header field when the + content size is known prior to sending the complete header section. + This will allow downstream recipients to measure transfer progress, + know when a received message is complete, and potentially reuse the + connection for additional requests.¶

+

+ Any Content-Length field value greater than or equal to zero is valid. + Since there is no predefined limit to the length of content, a + recipient MUST anticipate potentially large decimal numerals and + prevent parsing errors due to integer conversion overflows + or precision loss due to integer conversion + (Section 17.5).¶

+

+ Because Content-Length is used for message delimitation in HTTP/1.1, + its field value can impact how the message is parsed by downstream + recipients even when the immediate connection is not using HTTP/1.1. + If the message is forwarded by a downstream intermediary, a Content-Length + field value that is inconsistent with the received message framing might + cause a security failure due to request smuggling or response splitting.¶

+

+ As a result, a sender MUST NOT forward a message with a + Content-Length header field value that is known to be incorrect.¶

+

+ Likewise, a sender MUST NOT forward a message with a Content-Length + header field value that does not match the ABNF above, with one exception: + a recipient of a Content-Length header field value consisting of the same + decimal value repeated as a comma-separated list (e.g, + "Content-Length: 42, 42") MAY either reject the message as invalid or + replace that invalid field value with a single instance of the decimal + value, since this likely indicates that a duplicate was generated or + combined by an upstream message processor.¶

+
+
+
+
+

+8.7. Content-Location +

+ + + +

+ The "Content-Location" header field references a URI that can be used + as an identifier for a specific resource corresponding to the + representation in this message's content. + In other words, if one were to perform a GET request on this URI at the time + of this message's generation, then a 200 (OK) response would + contain the same representation that is enclosed as content in this message.¶

+ +
+
  Content-Location = absolute-URI / partial-URI
+
¶ +
+

+ The field value is either an absolute-URI or a + partial-URI. In the latter case (Section 4), + the referenced URI is relative to the target URI + ([URI], Section 5).¶

+

+ The Content-Location value is not a replacement for the target URI + (Section 7.1). It is representation metadata. + It has the same syntax and semantics as the header field of the same name + defined for MIME body parts in Section 4 of [RFC2557]. + However, its appearance in an HTTP message has some special implications + for HTTP recipients.¶

+

+ If Content-Location is included in a 2xx (Successful) + response message and its value refers (after conversion to absolute form) + to a URI that is the same as the target URI, then + the recipient MAY consider the content to be a current representation of + that resource at the time indicated by the message origination date. + For a GET (Section 9.3.1) or HEAD (Section 9.3.2) request, + this is the same as the default semantics when no Content-Location is + provided by the server. + For a state-changing request like PUT (Section 9.3.4) or + POST (Section 9.3.3), it implies that the server's response + contains the new representation of that resource, thereby distinguishing it + from representations that might only report about the action + (e.g., "It worked!"). + This allows authoring applications to update their local copies without + the need for a subsequent GET request.¶

+

+ If Content-Location is included in a 2xx (Successful) + response message and its field value refers to a URI that differs from the + target URI, then the origin server claims that the URI + is an identifier for a different resource corresponding to the enclosed + representation. Such a claim can only be trusted if both identifiers share + the same resource owner, which cannot be programmatically determined via + HTTP.¶

+
    +
  • For a response to a GET or HEAD request, this is an indication that the + target URI refers to a resource that is subject to content + negotiation and the Content-Location field value is a more specific + identifier for the selected representation.¶ +
    • +
  • For a 201 (Created) response to a state-changing method, + a Content-Location field value that is identical to the + Location field value indicates that this content is a + current representation of the newly created resource.¶ +
    • +
  • Otherwise, such a Content-Location indicates that this content is a + representation reporting on the requested action's status and that the + same report is available (for future access with GET) at the given URI. + For example, a purchase transaction made via a POST request might + include a receipt document as the content of the 200 (OK) + response; the Content-Location field value provides an identifier for + retrieving a copy of that same receipt in the future.¶ +
    • +
+

+ A user agent that sends Content-Location in a request message is stating + that its value refers to where the user agent originally obtained the + content of the enclosed representation (prior to any modifications made by + that user agent). In other words, the user agent is providing a back link + to the source of the original representation.¶

+

+ An origin server that receives a Content-Location field in a request + message MUST treat the information as transitory request context rather + than as metadata to be saved verbatim as part of the representation. + An origin server MAY use that context to guide in processing the + request or to save it for other uses, such as within source links or + versioning metadata. However, an origin server MUST NOT use such context + information to alter the request semantics.¶

+

+ For example, if a client makes a PUT request on a negotiated resource and + the origin server accepts that PUT (without redirection), then the new + state of that resource is expected to be consistent with the one + representation supplied in that PUT; the Content-Location cannot be used as + a form of reverse content selection identifier to update only one of the + negotiated representations. If the user agent had wanted the latter + semantics, it would have applied the PUT directly to the Content-Location + URI.¶

+
+
+
+
+

+8.8. Validator Fields +

+ + + +

+ Resource metadata is referred to as a "validator" if it + can be used within a precondition (Section 13.1) to + make a conditional request (Section 13). + Validator fields convey a current validator for the + selected representation + (Section 3.2).¶

+

+ In responses to safe requests, validator fields describe the selected + representation chosen by the origin server while handling the response. + Note that, depending on the method and status code semantics, the + selected representation for a given response is not + necessarily the same as the representation enclosed as response content.¶

+

+ In a successful response to a state-changing request, validator fields + describe the new representation that has replaced the prior + selected representation as a result of processing the + request.¶

+

+ For example, an ETag field in a 201 (Created) response + communicates the entity tag of the newly created resource's + representation, so that the entity tag can be used as a validator in + later conditional requests to prevent the "lost update" problem.¶

+

+ This specification defines two forms of metadata that are commonly used + to observe resource state and test for preconditions: modification dates + (Section 8.8.2) and opaque entity tags + (Section 8.8.3). + Additional metadata that reflects resource state + has been defined by various extensions of HTTP, such as Web Distributed + Authoring and Versioning [WEBDAV], that are beyond the + scope of this specification.¶

+
+
+

+8.8.1. Weak versus Strong +

+ + +

+ Validators come in two flavors: strong or weak. Weak validators are easy + to generate but are far less useful for comparisons. Strong validators + are ideal for comparisons but can be very difficult (and occasionally + impossible) to generate efficiently. Rather than impose that all forms + of resource adhere to the same strength of validator, HTTP exposes the + type of validator in use and imposes restrictions on when weak validators + can be used as preconditions.¶

+

+ A "strong validator" is representation metadata that changes value whenever + a change occurs to the representation data that would be observable in the + content of a 200 (OK) response to GET.¶

+

+ A strong validator might change for reasons other than a change to the + representation data, such as when a + semantically significant part of the representation metadata is changed + (e.g., Content-Type), but it is in the best interests of the + origin server to only change the value when it is necessary to invalidate + the stored responses held by remote caches and authoring tools.¶

+

+ Cache entries might persist for arbitrarily long periods, regardless + of expiration times. Thus, a cache might attempt to validate an + entry using a validator that it obtained in the distant past. + A strong validator is unique across all versions of all + representations associated with a particular resource over time. + However, there is no implication of uniqueness across representations + of different resources (i.e., the same strong validator might be + in use for representations of multiple resources at the same time + and does not imply that those representations are equivalent).¶

+

+ There are a variety of strong validators used in practice. The best are + based on strict revision control, wherein each change to a representation + always results in a unique node name and revision identifier being assigned + before the representation is made accessible to GET. + A collision-resistant hash + function applied to the representation data is also sufficient if the data + is available prior to the response header fields being sent and the digest + does not need to be recalculated every time a validation request is + received. However, if a resource has distinct representations that differ + only in their metadata, such as might occur with content negotiation over + media types that happen to share the same data format, then the origin + server needs to incorporate additional information in the validator to + distinguish those representations.¶

+

+ In contrast, a "weak validator" is representation metadata + that might not change for every change to the representation data. This + weakness might be due to limitations in how the value is calculated + (e.g., clock resolution), an inability to ensure uniqueness for all + possible representations of the resource, or a desire of the resource + owner to group representations by some self-determined set of + equivalency rather than unique sequences of data.¶

+

+ An origin server SHOULD change a weak entity tag whenever it + considers prior representations to be unacceptable as a substitute for + the current representation. In other words, a weak entity tag ought to + change whenever the origin server wants caches to invalidate old + responses.¶

+

+ For example, the representation of a weather report that changes in + content every second, based on dynamic measurements, might be grouped + into sets of equivalent representations (from the origin server's + perspective) with the same weak validator in order to allow cached + representations to be valid for a reasonable period of time (perhaps + adjusted dynamically based on server load or weather quality). + Likewise, a representation's modification time, if defined with only + one-second resolution, might be a weak validator if it is possible + for the representation to be modified twice during a single second and + retrieved between those modifications.¶

+

+ Likewise, a validator is weak if it is shared by two or more + representations of a given resource at the same time, unless those + representations have identical representation data. For example, if the + origin server sends the same validator for a representation with a gzip + content coding applied as it does for a representation with no content + coding, then that validator is weak. However, two simultaneous + representations might share the same strong validator if they differ only + in the representation metadata, such as when two different media types are + available for the same representation data.¶

+

+ Strong validators are usable for all conditional requests, including cache + validation, partial content ranges, and "lost update" avoidance. + Weak validators are only usable when the client does not require exact + equality with previously obtained representation data, such as when + validating a cache entry or limiting a web traversal to recent changes.¶

+
+
+
+
+

+8.8.2. Last-Modified +

+ + + +

+ The "Last-Modified" header field in a response provides a timestamp + indicating the date and time at which the origin server believes the + selected representation was last modified, as determined at the conclusion + of handling the request.¶

+ +
+
  Last-Modified = HTTP-date
+
¶ +
+

+ An example of its use is¶

+
+
Last-Modified: Tue, 15 Nov 1994 12:45:26 GMT
+
¶ +
+
+
+
+8.8.2.1. Generation +
+

+ An origin server SHOULD send Last-Modified for any selected + representation for which a last modification date can be reasonably + and consistently determined, since its use in conditional requests + and evaluating cache freshness ([CACHING]) can + substantially reduce unnecessary transfers and significantly + improve service availability and scalability.¶

+

+ A representation is typically the sum of many parts behind the + resource interface. The last-modified time would usually be + the most recent time that any of those parts were changed. + How that value is determined for any given resource is an + implementation detail beyond the scope of this specification.¶

+

+ An origin server SHOULD obtain the Last-Modified value of the + representation as close as possible to the time that it generates the + Date field value for its response. This allows a recipient to + make an accurate assessment of the representation's modification time, + especially if the representation changes near the time that the + response is generated.¶

+

+ An origin server with a clock (as defined in Section 5.6.7) + MUST NOT generate a Last-Modified date that is later than the + server's time of message origination + (Date, Section 6.6.1). + If the last modification time is derived from implementation-specific + metadata that evaluates to some time in the future, according to the + origin server's clock, then the origin server MUST replace that + value with the message origination date. This prevents a future + modification date from having an adverse impact on cache validation.¶

+

+ An origin server without a clock MUST NOT generate a Last-Modified + date for a response unless that date value was assigned to the resource + by some other system (presumably one with a clock).¶

+
+
+
+
+
+8.8.2.2. Comparison +
+

+ A Last-Modified time, when used as a validator in a request, is + implicitly weak unless it is possible to deduce that it is strong, + using the following rules:¶

+
    +
  • The validator is being compared by an origin server to the + actual current validator for the representation and,¶ +
    • +
  • That origin server reliably knows that the associated representation did + not change twice during the second covered by the presented + validator;¶ +
    • +
+

+ or¶

+
    +
  • The validator is about to be used by a client in an + If-Modified-Since, + If-Unmodified-Since, or If-Range header + field, because the client has a cache entry for the associated + representation, and¶ +
    • +
  • That cache entry includes a Date value which is + at least one second after the Last-Modified value and + the client has reason to believe that they were generated by the + same clock or that there is enough difference between the Last-Modified + and Date values to make clock synchronization issues unlikely;¶ +
    • +
+

+ or¶

+
    +
  • The validator is being compared by an intermediate cache to the + validator stored in its cache entry for the representation, and¶ +
    • +
  • That cache entry includes a Date value which is + at least one second after the Last-Modified value and + the cache has reason to believe that they were generated by the + same clock or that there is enough difference between the Last-Modified + and Date values to make clock synchronization issues unlikely.¶ +
    • +
+

+ This method relies on the fact that if two different responses were + sent by the origin server during the same second, but both had the + same Last-Modified time, then at least one of those responses would + have a Date value equal to its Last-Modified time.¶

+
+
+
+
+
+
+

+8.8.3. ETag +

+ + + + +

+ The "ETag" field in a response provides the current entity tag for + the selected representation, as determined at the conclusion of handling + the request. + An entity tag is an opaque validator for differentiating between + multiple representations of the same resource, regardless of whether + those multiple representations are due to resource state changes over + time, content negotiation resulting in multiple representations being + valid at the same time, or both. An entity tag consists of an opaque + quoted string, possibly prefixed by a weakness indicator.¶

+ + + + + +
+
  ETag       = entity-tag
+
+  entity-tag = [ weak ] opaque-tag
+  weak       = %s"W/"
+  opaque-tag = DQUOTE *etagc DQUOTE
+  etagc      = %x21 / %x23-7E / obs-text
+             ; VCHAR except double quotes, plus obs-text
+
¶ +
+ +

+ An entity tag can be more reliable for validation than a modification + date in situations where it is inconvenient to store modification + dates, where the one-second resolution of HTTP-date values is not + sufficient, or where modification dates are not consistently maintained.¶

+

+ Examples:¶

+
+
ETag: "xyzzy"
+ETag: W/"xyzzy"
+ETag: ""
+
¶ +
+

+ An entity tag can be either a weak or strong validator, with + strong being the default. If an origin server provides an entity tag + for a representation and the generation of that entity tag does not satisfy + all of the characteristics of a strong validator + (Section 8.8.1), then the origin server + MUST mark the entity tag as weak by prefixing its opaque value + with "W/" (case-sensitive).¶

+

+ A sender MAY send the ETag field in a trailer section (see + Section 6.5). However, since trailers are often + ignored, it is preferable to send ETag as a header field unless the + entity tag is generated while sending the content.¶

+
+
+
+8.8.3.1. Generation +
+

+ The principle behind entity tags is that only the service author + knows the implementation of a resource well enough to select the + most accurate and efficient validation mechanism for that resource, + and that any such mechanism can be mapped to a simple sequence of + octets for easy comparison. Since the value is opaque, there is no + need for the client to be aware of how each entity tag is constructed.¶

+

+ For example, a resource that has implementation-specific versioning + applied to all changes might use an internal revision number, perhaps + combined with a variance identifier for content negotiation, to + accurately differentiate between representations. + Other implementations might use a collision-resistant hash of + representation content, a combination of various file attributes, or + a modification timestamp that has sub-second resolution.¶

+

+ An origin server SHOULD send an ETag for any selected representation + for which detection of changes can be reasonably and consistently + determined, since the entity tag's use in conditional requests and + evaluating cache freshness ([CACHING]) can + substantially reduce unnecessary transfers and significantly + improve service availability, scalability, and reliability.¶

+
+
+
+
+
+8.8.3.2. Comparison +
+

+ There are two entity tag comparison functions, depending on whether or not + the comparison context allows the use of weak validators:¶

+
+
+ "Strong comparison": +
+
+ two entity tags are equivalent if both are not weak and their opaque-tags + match character-by-character.¶ +
+
+
+ "Weak comparison": +
+
+ two entity tags are equivalent if their opaque-tags match + character-by-character, regardless of either or both being tagged as "weak".¶ +
+
+
+

+ The example below shows the results for a set of entity tag pairs and both + the weak and strong comparison function results:¶

+ + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +
Table 3
ETag 1ETag 2Strong ComparisonWeak Comparison
W/"1"W/"1"no matchmatch
W/"1"W/"2"no matchno match
W/"1""1"no matchmatch
"1""1"matchmatch
+
+
+
+
+
+8.8.3.3. Example: Entity Tags Varying on Content-Negotiated Resources +
+

+ Consider a resource that is subject to content negotiation + (Section 12), and where the representations sent in response to + a GET request vary based on the Accept-Encoding request + header field (Section 12.5.3):¶

+

+ >> Request:¶

+
+
GET /index HTTP/1.1
+Host: www.example.com
+Accept-Encoding: gzip
+
+
¶ +
+

+ In this case, the response might or might not use the gzip content coding. + If it does not, the response might look like:¶

+

+ >> Response:¶

+
+
HTTP/1.1 200 OK
+Date: Fri, 26 Mar 2010 00:05:00 GMT
+ETag: "123-a"
+Content-Length: 70
+Vary: Accept-Encoding
+Content-Type: text/plain
+
+Hello World!
+Hello World!
+Hello World!
+Hello World!
+Hello World!
+
¶ +
+

+ An alternative representation that does use gzip content coding would be:¶

+

+ >> Response:¶

+
+
HTTP/1.1 200 OK
+Date: Fri, 26 Mar 2010 00:05:00 GMT
+ETag: "123-b"
+Content-Length: 43
+Vary: Accept-Encoding
+Content-Type: text/plain
+Content-Encoding: gzip
+
+...binary data...
¶ +
+ +
+
+
+
+
+
+
+
+
+
+

+9. Methods +

+
+
+

+9.1. Overview +

+

+ The request method token is the primary source of request semantics; + it indicates the purpose for which the client has made this request + and what is expected by the client as a successful result.¶

+

+ The request method's semantics might be further specialized by the + semantics of some header fields when present in a request + if those additional semantics do not conflict with the method. + For example, a client can send conditional request header fields + (Section 13.1) to make the requested + action conditional on the current state of the target resource.¶

+

+ HTTP is designed to be usable as an interface to distributed + object systems. The request method invokes an action to be applied to + a target resource in much the same way that a remote + method invocation can be sent to an identified object.¶

+ +
+
  method = token
+
¶ +
+

+ The method token is case-sensitive because it might be used as a gateway + to object-based systems with case-sensitive method names. By convention, + standardized methods are defined in all-uppercase US-ASCII letters.¶

+

+ Unlike distributed objects, the standardized request methods in HTTP are + not resource-specific, since uniform interfaces provide for better + visibility and reuse in network-based systems [REST]. + Once defined, a standardized method ought to have the same semantics when + applied to any resource, though each resource determines for itself + whether those semantics are implemented or allowed.¶

+

+ This specification defines a number of standardized methods that are + commonly used in HTTP, as outlined by the following table.¶

+
+ + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +
Table 4
Method NameDescriptionSection
GETTransfer a current representation of the target resource. + 9.3.1 +
HEADSame as GET, but do not transfer the response content. + 9.3.2 +
POSTPerform resource-specific processing on the request content. + 9.3.3 +
PUTReplace all current representations of the target resource with + the request content. + 9.3.4 +
DELETERemove all current representations of the target resource. + 9.3.5 +
CONNECTEstablish a tunnel to the server identified by the target resource. + 9.3.6 +
OPTIONSDescribe the communication options for the target resource. + 9.3.7 +
TRACEPerform a message loop-back test along the path to the target resource. + 9.3.8 +
+
+

+ All general-purpose servers MUST support the methods GET and HEAD. + All other methods are OPTIONAL.¶

+

+ The set of methods allowed by a target resource can be listed in an + Allow header field (Section 10.2.1). + However, the set of allowed methods can change dynamically. + An origin server that receives a request method that is unrecognized or + not implemented SHOULD respond with the + 501 (Not Implemented) status code. + An origin server that receives a request method that is recognized and + implemented, but not allowed for the target resource, SHOULD respond + with the 405 (Method Not Allowed) status code.¶

+

+ Additional methods, outside the scope of this specification, have been + specified for use in HTTP. All such methods ought to be registered + within the "Hypertext Transfer Protocol (HTTP) Method Registry", + as described in Section 16.1.¶

+
+
+
+
+

+9.2. Common Method Properties +

+
+
+

+9.2.1. Safe Methods +

+ +

+ Request methods are considered "safe" if + their defined semantics are essentially read-only; i.e., the client does + not request, and does not expect, any state change on the origin server + as a result of applying a safe method to a target resource. Likewise, + reasonable use of a safe method is not expected to cause any harm, + loss of property, or unusual burden on the origin server.¶

+

+ This definition of safe methods does not prevent an implementation from + including behavior that is potentially harmful, that is not entirely read-only, + or that causes side effects while invoking a safe method. What is + important, however, is that the client did not request that additional + behavior and cannot be held accountable for it. For example, + most servers append request information to access log files at the + completion of every response, regardless of the method, and that is + considered safe even though the log storage might become full and cause + the server to fail. Likewise, a safe request initiated by selecting an + advertisement on the Web will often have the side effect of charging an + advertising account.¶

+

+ Of the request methods defined by this specification, the + GET, HEAD, OPTIONS, and + TRACE methods are defined to be safe.¶

+

+ The purpose of distinguishing between safe and unsafe methods is to + allow automated retrieval processes (spiders) and cache performance + optimization (pre-fetching) to work without fear of causing harm. + In addition, it allows a user agent to apply appropriate constraints + on the automated use of unsafe methods when processing potentially + untrusted content.¶

+

+ A user agent SHOULD distinguish between safe and unsafe methods when + presenting potential actions to a user, such that the user can be made + aware of an unsafe action before it is requested.¶

+

+ When a resource is constructed such that parameters within the target URI + have the effect of selecting an action, it is the resource + owner's responsibility to ensure that the action is consistent with the + request method semantics. + For example, it is common for Web-based content editing software + to use actions within query parameters, such as "page?do=delete". + If the purpose of such a resource is to perform an unsafe action, then + the resource owner MUST disable or disallow that action when it is + accessed using a safe request method. Failure to do so will result in + unfortunate side effects when automated processes perform a GET on + every URI reference for the sake of link maintenance, pre-fetching, + building a search index, etc.¶

+
+
+
+
+

+9.2.2. Idempotent Methods +

+ +

+ A request method is considered "idempotent" + if the intended effect on the server of multiple identical requests with + that method is the same as the effect for a single such request. + Of the request methods defined by this + specification, PUT, DELETE, and safe request + methods are idempotent.¶

+

+ Like the definition of safe, the idempotent property only applies to + what has been requested by the user; a server is free to log each request + separately, retain a revision control history, or implement other + non-idempotent side effects for each idempotent request.¶

+

+ Idempotent methods are distinguished because the request can be repeated + automatically if a communication failure occurs before the client is + able to read the server's response. For example, if a client sends a PUT + request and the underlying connection is closed before any response is + received, then the client can establish a new connection and retry the + idempotent request. It knows that repeating the request will have + the same intended effect, even if the original request succeeded, though + the response might differ.¶

+

+ A client SHOULD NOT automatically retry a request with a non-idempotent + method unless it has some means to know that the request semantics are + actually idempotent, regardless of the method, or some means to detect that + the original request was never applied.¶

+

+ For example, a user agent can repeat a POST request automatically if it + knows (through design or configuration) that the request is safe for that + resource. Likewise, a user agent designed specifically to operate on + a version control repository might be able to recover from partial failure + conditions by checking the target resource revision(s) after a failed + connection, reverting or fixing any changes that were partially applied, + and then automatically retrying the requests that failed.¶

+

+ Some clients take a riskier approach and attempt to guess when an + automatic retry is possible. For example, a client might automatically + retry a POST request if the underlying transport connection closed before + any part of a response is received, particularly if an idle persistent + connection was used.¶

+

+ A proxy MUST NOT automatically retry non-idempotent requests. + A client SHOULD NOT automatically retry a failed automatic retry.¶

+
+
+
+
+

+9.2.3. Methods and Caching +

+

+ For a cache to store and use a response, the associated method needs to + explicitly allow caching and to detail under what conditions a response can + be used to satisfy subsequent requests; a method definition that does not + do so cannot be cached. For additional requirements see [CACHING].¶

+

+ This specification defines caching semantics for GET, HEAD, and POST, + although the overwhelming majority of cache implementations only support + GET and HEAD.¶

+
+
+
+
+
+
+

+9.3. Method Definitions +

+
+
+

+9.3.1. GET +

+ + +

+ The GET method requests transfer of a current + selected representation for the + target resource. + A successful response reflects the quality of "sameness" identified by + the target URI (Section 1.2.2 of [URI]). Hence, + retrieving identifiable information via HTTP is usually performed by + making a GET request on an identifier associated with the potential for + providing that information in a 200 (OK) response.¶

+

+ GET is the primary mechanism of information retrieval and the focus of + almost all performance optimizations. Applications that produce a URI for + each important resource can benefit from those optimizations while enabling + their reuse by other applications, creating a network effect that promotes + further expansion of the Web.¶

+

+ It is tempting to think of resource identifiers as remote file system + pathnames and of representations as being a copy of the contents of such + files. In fact, that is how many resources are implemented (see + Section 17.3 for related security considerations). + However, there are no such limitations in practice.¶

+

+ The HTTP interface for + a resource is just as likely to be implemented as a tree of content + objects, a programmatic view on various database records, or a gateway to + other information systems. Even when the URI mapping mechanism is tied to a + file system, an origin server might be configured to execute the files with + the request as input and send the output as the representation rather than + transfer the files directly. Regardless, only the origin server needs to + know how each resource identifier corresponds to an implementation + and how that implementation manages to select and send a current + representation of the target resource.¶

+

+ A client can alter the semantics of GET to be a "range request", requesting + transfer of only some part(s) of the selected representation, by sending a + Range header field in the request (Section 14.2).¶

+

+ Although request message framing is independent of the method used, + content received in a GET request has no generally defined semantics, + cannot alter the meaning or target of the request, and might lead some + implementations to reject the request and close the connection because of + its potential as a request smuggling attack + (Section 11.2 of [HTTP/1.1]). + A client SHOULD NOT generate content in a GET request unless it is + made directly to an origin server that has previously indicated, + in or out of band, that such a request has a purpose and will be adequately + supported. An origin server SHOULD NOT rely on private agreements to + receive content, since participants in HTTP communication are often + unaware of intermediaries along the request chain.¶

+

+ The response to a GET request is cacheable; a cache MAY use it to satisfy + subsequent GET and HEAD requests unless otherwise indicated by the + Cache-Control header field (Section 5.2 of [CACHING]).¶

+

+ When information retrieval is performed with a mechanism that constructs a + target URI from user-provided information, such as the query fields of a + form using GET, potentially sensitive data might be provided that would not + be appropriate for disclosure within a URI + (see Section 17.9). In some cases, the + data can be filtered or transformed such that it would not reveal such + information. In others, particularly when there is no benefit from caching + a response, using the POST method (Section 9.3.3) instead of GET + can transmit such information in the request content rather than within + the target URI.¶

+
+
+ +
+
+

+9.3.3. POST +

+ + +

+ The POST method requests that the target resource process + the representation enclosed in the request according to the resource's own + specific semantics. For example, POST is used for the following functions + (among others):¶

+
    +
  • Providing a block of data, such as the fields entered into an HTML + form, to a data-handling process;¶ +
    • +
  • Posting a message to a bulletin board, newsgroup, mailing list, blog, + or similar group of articles;¶ +
    • +
  • Creating a new resource that has yet to be identified by the origin + server; and¶ +
    • +
  • Appending data to a resource's existing representation(s).¶ +
    • +
+

+ An origin server indicates response semantics by choosing an appropriate + status code depending on the result of processing the POST request; + almost all of the status codes defined by this specification could be + received in a response to POST (the exceptions being 206 (Partial Content), + 304 (Not Modified), and 416 (Range Not Satisfiable)).¶

+

+ If one or more resources has been created on the origin server as a result + of successfully processing a POST request, the origin server SHOULD send + a 201 (Created) response containing a Location + header field that provides an identifier for the primary resource created + (Section 10.2.2) and a representation that describes the + status of the request while referring to the new resource(s).¶

+

+ Responses to POST requests are only cacheable when they include explicit + freshness information (see Section 4.2.1 of [CACHING]) and a + Content-Location header field that has the same value as + the POST's target URI (Section 8.7). A cached POST response can be reused + to satisfy a later GET or HEAD request. In contrast, a POST request cannot + be satisfied by a cached POST response because POST is potentially unsafe; + see Section 4 of [CACHING].¶

+

+ If the result of processing a POST would be equivalent to a representation + of an existing resource, an origin server MAY redirect the user agent to + that resource by sending a 303 (See Other) response with the + existing resource's identifier in the Location field. + This has the benefits of providing the user agent a resource identifier + and transferring the representation via a method more amenable to shared + caching, though at the cost of an extra request if the user agent does not + already have the representation cached.¶

+
+
+
+
+

+9.3.4. PUT +

+ + +

+ The PUT method requests that the state of the target resource + be created or replaced with the state defined by the representation + enclosed in the request message content. A successful PUT of a given + representation would suggest that a subsequent GET on that same target + resource will result in an equivalent representation being sent in + a 200 (OK) response. However, there is no guarantee that + such a state change will be observable, since the target resource might be + acted upon by other user agents in parallel, or might be subject to dynamic + processing by the origin server, before any subsequent GET is received. + A successful response only implies that the user agent's intent was + achieved at the time of its processing by the origin server.¶

+

+ If the target resource does not have a current representation and + the PUT successfully creates one, then the origin server MUST inform + the user agent by sending a 201 (Created) response. If the + target resource does have a current representation and that representation is + successfully modified in accordance with the state of the enclosed + representation, then the origin server MUST send either a + 200 (OK) or a 204 (No Content) response to + indicate successful completion of the request.¶

+

+ An origin server SHOULD verify that the PUT representation is consistent + with its configured constraints for the target resource. For example, if + an origin server determines a resource's representation metadata based on + the URI, then the origin server needs to ensure that the content received + in a successful PUT request is consistent with that metadata. When a PUT + representation is inconsistent with the target resource, the origin + server SHOULD either make them consistent, by transforming the + representation or changing the resource configuration, or respond + with an appropriate error message containing sufficient information + to explain why the representation is unsuitable. The + 409 (Conflict) or 415 (Unsupported Media Type) + status codes are suggested, with the latter being specific to constraints on + Content-Type values.¶

+

+ For example, if the target resource is configured to always have a + Content-Type of "text/html" and the representation being PUT + has a Content-Type of "image/jpeg", the origin server ought to do one of:¶

+
    +
  1. reconfigure the target resource to reflect the new media type;¶ +
    2. +
  2. transform the PUT representation to a format consistent with that + of the resource before saving it as the new resource state; or,¶ +
    3. +
  3. reject the request with a 415 (Unsupported Media Type) + response indicating that the target resource is limited to "text/html", + perhaps including a link to a different resource that would be a + suitable target for the new representation.¶ +
    4. +
+

+ HTTP does not define exactly how a PUT method affects the state + of an origin server beyond what can be expressed by the intent of + the user agent request and the semantics of the origin server response. + It does not define what a resource might be, in any sense of that + word, beyond the interface provided via HTTP. It does not define + how resource state is "stored", nor how such storage might change + as a result of a change in resource state, nor how the origin server + translates resource state into representations. Generally speaking, + all implementation details behind the resource interface are + intentionally hidden by the server.¶

+

+ This extends to how header and trailer fields are stored; while common + header fields like Content-Type will typically be stored + and returned upon subsequent GET requests, header and trailer field + handling is specific to the resource that received the request. As a result, + an origin server SHOULD ignore unrecognized header and trailer fields + received in a PUT request (i.e., not save them as part of the resource + state).¶

+

+ An origin server MUST NOT send a validator field + (Section 8.8), such as an ETag or + Last-Modified field, in a successful response to PUT unless + the request's representation data was saved without any transformation + applied to the content (i.e., the resource's new representation data is + identical to the content received in the PUT request) and the + validator field value reflects the new representation. + This requirement allows a user agent to know when the representation it + sent (and retains in memory) is the result of the PUT, and thus it doesn't + need to be retrieved again from the origin server. The new validator(s) + received in the response can be used for future conditional requests in + order to prevent accidental overwrites (Section 13.1).¶

+

+ The fundamental difference between the POST and PUT methods is + highlighted by the different intent for the enclosed representation. + The target resource in a POST request is intended to handle the + enclosed representation according to the resource's own semantics, + whereas the enclosed representation in a PUT request is defined as + replacing the state of the target resource. Hence, the intent of PUT is + idempotent and visible to intermediaries, even though the exact effect is + only known by the origin server.¶

+

+ Proper interpretation of a PUT request presumes that the user agent knows + which target resource is desired. A service that selects a proper URI on + behalf of the client, after receiving a state-changing request, SHOULD be + implemented using the POST method rather than PUT. If the origin server + will not make the requested PUT state change to the target resource and + instead wishes to have it applied to a different resource, such as when the + resource has been moved to a different URI, then the origin server MUST + send an appropriate 3xx (Redirection) response; the + user agent MAY then make its own decision regarding whether or not to + redirect the request.¶

+

+ A PUT request applied to the target resource can have side effects + on other resources. For example, an article might have a URI for + identifying "the current version" (a resource) that is separate + from the URIs identifying each particular version (different + resources that at one point shared the same state as the current version + resource). A successful PUT request on "the current version" URI might + therefore create a new version resource in addition to changing the + state of the target resource, and might also cause links to be added + between the related resources.¶

+

+ Some origin servers support use of the Content-Range header + field (Section 14.4) as a request modifier to + perform a partial PUT, as described in Section 14.5.¶

+

+ Responses to the PUT method are not cacheable. If a successful PUT request + passes through a cache that has one or more stored responses for the + target URI, those stored responses will be invalidated + (see Section 4.4 of [CACHING]).¶

+
+
+
+
+

+9.3.5. DELETE +

+ + +

+ The DELETE method requests that the origin server remove the association + between the target resource and its current functionality. + In effect, this method is similar to the "rm" command in UNIX: it expresses a + deletion operation on the URI mapping of the origin server rather than an + expectation that the previously associated information be deleted.¶

+

+ If the target resource has one or more current representations, they might + or might not be destroyed by the origin server, and the associated storage + might or might not be reclaimed, depending entirely on the nature of the + resource and its implementation by the origin server (which are beyond the + scope of this specification). Likewise, other implementation aspects of a + resource might need to be deactivated or archived as a result of a DELETE, + such as database or gateway connections. In general, it is assumed that the + origin server will only allow DELETE on resources for which it has a + prescribed mechanism for accomplishing the deletion.¶

+

+ Relatively few resources allow the DELETE method -- its primary use + is for remote authoring environments, where the user has some direction + regarding its effect. For example, a resource that was previously created + using a PUT request, or identified via the Location header field after a + 201 (Created) response to a POST request, might allow a + corresponding DELETE request to undo those actions. Similarly, custom + user agent implementations that implement an authoring function, such as + revision control clients using HTTP for remote operations, might use + DELETE based on an assumption that the server's URI space has been crafted + to correspond to a version repository.¶

+

+ If a DELETE method is successfully applied, the origin server SHOULD send¶

+
    +
  • a 202 (Accepted) status code if the action will likely succeed but + has not yet been enacted,¶ +
    • +
  • a 204 (No Content) status code if the action has been + enacted and no further information is to be supplied, or¶ +
    • +
  • a 200 (OK) status code if the action has been enacted and + the response message includes a representation describing the status.¶ +
    • +
+

+ Although request message framing is independent of the method used, + content received in a DELETE request has no generally defined semantics, + cannot alter the meaning or target of the request, and might lead some + implementations to reject the request and close the connection because of + its potential as a request smuggling attack + (Section 11.2 of [HTTP/1.1]). + A client SHOULD NOT generate content in a DELETE request unless it is + made directly to an origin server that has previously indicated, + in or out of band, that such a request has a purpose and will be adequately + supported. An origin server SHOULD NOT rely on private agreements to + receive content, since participants in HTTP communication are often + unaware of intermediaries along the request chain.¶

+

+ Responses to the DELETE method are not cacheable. If a successful DELETE + request passes through a cache that has one or more stored responses for + the target URI, those stored responses will be invalidated (see + Section 4.4 of [CACHING]).¶

+
+
+
+
+

+9.3.6. CONNECT +

+ + +

+ The CONNECT method requests that the recipient establish a tunnel to the + destination origin server identified by the request target and, if + successful, thereafter restrict its behavior to blind forwarding of + data, in both directions, until the tunnel is closed. + Tunnels are commonly used to create an end-to-end virtual connection, + through one or more proxies, which can then be secured using TLS + (Transport Layer Security, [TLS13]).¶

+

+ CONNECT uses a special form of request target, unique to this method, + consisting of only the host and port number of the tunnel destination, + separated by a colon. There is no default port; a client MUST send the + port number even if the CONNECT request is based on a URI reference that + contains an authority component with an elided port + (Section 4.1). For example,¶

+
+
CONNECT server.example.com:80 HTTP/1.1
+Host: server.example.com
+
+
¶ +
+

+ A server MUST reject a CONNECT request that targets an empty or invalid + port number, typically by responding with a 400 (Bad Request) status code.¶

+

+ Because CONNECT changes the request/response nature of an HTTP connection, + specific HTTP versions might have different ways of mapping its semantics + into the protocol's wire format.¶

+

+ CONNECT is intended for use in requests to a proxy. + The recipient can establish a tunnel either by directly connecting to + the server identified by the request target or, if configured to use + another proxy, by forwarding the CONNECT request to the next inbound proxy. + An origin server MAY accept a CONNECT request, but most origin servers + do not implement CONNECT.¶

+

+ Any 2xx (Successful) response indicates + that the sender (and all inbound proxies) will switch to tunnel mode + immediately after the response header section; data received after that + header section is from the server identified by the request target. + Any response other than a successful response indicates that the tunnel + has not yet been formed.¶

+

+ A tunnel is closed when a tunnel intermediary detects that either side + has closed its connection: the intermediary MUST attempt to send any + outstanding data that came from the closed side to the other side, close + both connections, and then discard any remaining data left undelivered.¶

+

+ Proxy authentication might be used to establish the + authority to create a tunnel. For example,¶

+
+
CONNECT server.example.com:443 HTTP/1.1
+Host: server.example.com:443
+Proxy-Authorization: basic aGVsbG86d29ybGQ=
+
+
¶ +
+

+ There are significant risks in establishing a tunnel to arbitrary servers, + particularly when the destination is a well-known or reserved TCP port that + is not intended for Web traffic. For example, a CONNECT to + "example.com:25" would suggest that the proxy connect to the reserved + port for SMTP traffic; if allowed, that could trick the proxy into + relaying spam email. Proxies that support CONNECT SHOULD restrict its + use to a limited set of known ports or a configurable list of safe + request targets.¶

+

+ A server MUST NOT send any Transfer-Encoding or + Content-Length header fields in a + 2xx (Successful) response to CONNECT. + A client MUST ignore any Content-Length or Transfer-Encoding header + fields received in a successful response to CONNECT.¶

+

+ A CONNECT request message does not have content. The interpretation of + data sent after the header section of the CONNECT request message is + specific to the version of HTTP in use.¶

+

+ Responses to the CONNECT method are not cacheable.¶

+
+
+
+
+

+9.3.7. OPTIONS +

+ + +

+ The OPTIONS method requests information about the communication options + available for the target resource, at either the origin server or an + intervening intermediary. This method allows a client to determine the + options and/or requirements associated with a resource, or the capabilities + of a server, without implying a resource action.¶

+

+ An OPTIONS request with an asterisk ("*") as the request target + (Section 7.1) applies to the server in general rather than to a + specific resource. Since a server's communication options typically depend + on the resource, the "*" request is only useful as a "ping" or "no-op" + type of method; it does nothing beyond allowing the client to test + the capabilities of the server. For example, this can be used to test + a proxy for HTTP/1.1 conformance (or lack thereof).¶

+

+ If the request target is not an asterisk, the OPTIONS request applies + to the options that are available when communicating with the target + resource.¶

+

+ A server generating a successful response to OPTIONS SHOULD send any + header that might indicate optional features implemented by the + server and applicable to the target resource (e.g., Allow), + including potential extensions not defined by this specification. + The response content, if any, might also describe the communication options + in a machine or human-readable representation. A standard format for such a + representation is not defined by this specification, but might be defined by + future extensions to HTTP.¶

+

+ A client MAY send a Max-Forwards header field in an + OPTIONS request to target a specific recipient in the request chain (see + Section 7.6.2). A proxy MUST NOT generate a + Max-Forwards header field while forwarding a request unless that request + was received with a Max-Forwards field.¶

+

+ A client that generates an OPTIONS request containing content + MUST send a valid Content-Type header field describing + the representation media type. Note that this specification does not define + any use for such content.¶

+

+ Responses to the OPTIONS method are not cacheable.¶

+
+
+
+
+

+9.3.8. TRACE +

+ + +

+ The TRACE method requests a remote, application-level loop-back of the + request message. The final recipient of the request SHOULD reflect the + message received, excluding some fields described below, back to the client + as the content of a 200 (OK) response. The "message/http" + format (Section 10.1 of [HTTP/1.1]) is one way to do so. + The final recipient is either the origin server or the first server to + receive a Max-Forwards value of zero (0) in the request + (Section 7.6.2).¶

+

+ A client MUST NOT generate fields in a TRACE request containing + sensitive data that might be disclosed by the response. For example, it + would be foolish for a user agent to send stored user credentials + (Section 11) or cookies [COOKIE] in a TRACE + request. The final recipient of the request SHOULD exclude any request + fields that are likely to contain sensitive data when that recipient + generates the response content.¶

+

+ TRACE allows the client to see what is being received at the other + end of the request chain and use that data for testing or diagnostic + information. The value of the Via header field (Section 7.6.3) + is of particular interest, since it acts as a trace of the request chain. + Use of the Max-Forwards header field allows the client to + limit the length of the request chain, which is useful for testing a chain + of proxies forwarding messages in an infinite loop.¶

+

+ A client MUST NOT send content in a TRACE request.¶

+

+ Responses to the TRACE method are not cacheable.¶

+
+
+
+
+
+
+
+
+

+10. Message Context +

+
+
+

+10.1. Request Context Fields +

+

+ The request header fields below provide additional information about the + request context, including information about the user, user agent, and + resource behind the request.¶

+
+
+

+10.1.1. Expect +

+ + + + +

+ The "Expect" header field in a request indicates a certain set of + behaviors (expectations) that need to be supported by the server in + order to properly handle this request.¶

+ +
+
  Expect =      #expectation
+  expectation = token [ "=" ( token / quoted-string ) parameters ]
+
¶ +
+

+ The Expect field value is case-insensitive.¶

+

+ The only expectation defined by this specification is "100-continue" + (with no defined parameters).¶

+

+ A server that receives an Expect field value containing a member other than + 100-continue + MAY respond with a + 417 (Expectation Failed) status code to indicate that the + unexpected expectation cannot be met.¶

+

+ A "100-continue" expectation informs recipients that the + client is about to send (presumably large) content in this request + and wishes to receive a 100 (Continue) interim response if + the method, target URI, and header fields are not sufficient to cause an immediate + success, redirect, or error response. This allows the client to wait for an + indication that it is worthwhile to send the content before actually + doing so, which can improve efficiency when the data is huge or + when the client anticipates that an error is likely (e.g., when sending a + state-changing method, for the first time, without previously verified + authentication credentials).¶

+

+ For example, a request that begins with¶

+
+
PUT /somewhere/fun HTTP/1.1
+Host: origin.example.com
+Content-Type: video/h264
+Content-Length: 1234567890987
+Expect: 100-continue
+
+
¶ +
+

+ allows the origin server to immediately respond with an error message, such + as 401 (Unauthorized) or 405 (Method Not Allowed), + before the client starts filling the pipes with an unnecessary data + transfer.¶

+

+ Requirements for clients:¶

+
    +
  • + A client MUST NOT generate a 100-continue expectation in a request that + does not include content.¶ +
    • +
  • + A client that will wait for a 100 (Continue) response + before sending the request content MUST send an + Expect header field containing a 100-continue expectation.¶ +
    • +
  • + A client that sends a 100-continue expectation is not required to wait + for any specific length of time; such a client MAY proceed to send the + content even if it has not yet received a response. Furthermore, + since 100 (Continue) responses cannot be sent through an + HTTP/1.0 intermediary, such a client SHOULD NOT wait for an indefinite + period before sending the content.¶ +
    • +
  • + A client that receives a 417 (Expectation Failed) status + code in response to a request containing a 100-continue expectation + SHOULD repeat that request without a 100-continue expectation, since + the 417 response merely indicates that the response chain does not + support expectations (e.g., it passes through an HTTP/1.0 server).¶ +
    • +
+

+ Requirements for servers:¶

+
    +
  • + A server that receives a 100-continue expectation in an HTTP/1.0 request + MUST ignore that expectation.¶ +
    • +
  • + A server MAY omit sending a 100 (Continue) response if + it has already received some or all of the content for the + corresponding request, or if the framing indicates that there is no + content.¶ +
    • +
  • + A server that sends a 100 (Continue) response MUST + ultimately send a final status code, once it receives and processes the + request content, unless the connection is closed prematurely.¶ +
    • +
  • + A server that responds with a final status code before reading the + entire request content SHOULD indicate whether it intends to + close the connection (e.g., see Section 9.6 of [HTTP/1.1]) or + continue reading the request content.¶ +
    • +
+

+ Upon receiving an HTTP/1.1 (or later) request that has a method, target URI, + and complete header section that contains a 100-continue expectation and + an indication that request content will follow, an origin server MUST + send either:¶

+
    +
  • an immediate response with a final status code, if that status can be + determined by examining just the method, target URI, and header fields, or¶ +
    • +
  • an immediate 100 (Continue) response to encourage the client + to send the request content.¶ +
    • +
+

+ The origin server MUST NOT wait for the content + before sending the 100 (Continue) response.¶

+

+ Upon receiving an HTTP/1.1 (or later) request that has a method, target URI, + and complete header section that contains a 100-continue expectation and + indicates a request content will follow, a proxy MUST either:¶

+
    +
  • send an immediate + response with a final status code, if that status can be determined by + examining just the method, target URI, and header fields, or¶ +
    • +
  • forward the request toward the origin server by sending a corresponding + request-line and header section to the next inbound server.¶ +
    • +
+

+ If the proxy believes (from configuration or past interaction) that the + next inbound server only supports HTTP/1.0, the proxy MAY generate an + immediate 100 (Continue) response to encourage the client to + begin sending the content.¶

+
+
+
+
+

+10.1.2. From +

+ + + +

+ The "From" header field contains an Internet email address for a human + user who controls the requesting user agent. The address ought to be + machine-usable, as defined by "mailbox" + in Section 3.4 of [RFC5322]:¶

+ +
+
  From    = mailbox
+
+  mailbox = <mailbox, see [RFC5322], Section 3.4>
+
¶ +
+

+ An example is:¶

+
+
From: spider-admin@example.org
+
¶ +
+

+ The From header field is rarely sent by non-robotic user agents. + A user agent SHOULD NOT send a From header field without explicit + configuration by the user, since that might conflict with the user's + privacy interests or their site's security policy.¶

+

+ A robotic user agent SHOULD send a valid From header field so that the + person responsible for running the robot can be contacted if problems + occur on servers, such as if the robot is sending excessive, unwanted, + or invalid requests.¶

+

+ A server SHOULD NOT use the From header field for access control or + authentication, since its value is expected to be visible to anyone + receiving or observing the request and is often recorded within logfiles + and error reports without any expectation of privacy.¶

+
+
+
+
+

+10.1.3. Referer +

+ + + +

+ The "Referer" [sic] header field allows the user agent to specify a URI + reference for the resource from which the target URI was + obtained (i.e., the "referrer", though the field name is misspelled). + A user agent MUST NOT include the fragment and userinfo components + of the URI reference [URI], if any, when generating the + Referer field value.¶

+ +
+
  Referer = absolute-URI / partial-URI
+
¶ +
+

+ The field value is either an absolute-URI or a + partial-URI. In the latter case (Section 4), + the referenced URI is relative to the target URI + ([URI], Section 5).¶

+

+ The Referer header field allows servers to generate back-links to other + resources for simple analytics, logging, optimized caching, etc. It also + allows obsolete or mistyped links to be found for maintenance. Some servers + use the Referer header field as a means of denying links from other sites + (so-called "deep linking") or restricting cross-site request forgery (CSRF), + but not all requests contain it.¶

+

+ Example:¶

+
+
Referer: http://www.example.org/hypertext/Overview.html
+
¶ +
+

+ If the target URI was obtained from a source that does not have its own + URI (e.g., input from the user keyboard, or an entry within the user's + bookmarks/favorites), the user agent MUST either exclude the Referer header field + or send it with a value of "about:blank".¶

+

+ The Referer header field value need not convey the full URI of the referring + resource; a user agent MAY truncate parts other than the referring origin.¶

+

+ The Referer header field has the potential to reveal information about the request + context or browsing history of the user, which is a privacy concern if the + referring resource's identifier reveals personal information (such as an + account name) or a resource that is supposed to be confidential (such as + behind a firewall or internal to a secured service). Most general-purpose + user agents do not send the Referer header field when the referring + resource is a local "file" or "data" URI. A user agent SHOULD NOT send a + Referer header field if the referring resource was accessed with + a secure protocol and the request target has an origin differing from that + of the referring resource, unless the referring resource explicitly allows + Referer to be sent. A user agent MUST NOT send a + Referer header field in an unsecured HTTP request if the + referring resource was accessed with a secure protocol. + See Section 17.9 for additional + security considerations.¶

+

+ Some intermediaries have been known to indiscriminately remove Referer + header fields from outgoing requests. This has the unfortunate side effect + of interfering with protection against CSRF attacks, which can be far + more harmful to their users. Intermediaries and user agent extensions that + wish to limit information disclosure in Referer ought to restrict their + changes to specific edits, such as replacing internal domain names with + pseudonyms or truncating the query and/or path components. + An intermediary SHOULD NOT modify or delete the Referer header field when + the field value shares the same scheme and host as the target URI.¶

+
+
+
+
+

+10.1.4. TE +

+ + + +

+ The "TE" header field describes capabilities of the client with regard to + transfer codings and trailer sections.¶

+

+ As described in Section 6.5, + a TE field with a "trailers" member sent in a request indicates that the + client will not discard trailer fields.¶

+

+ TE is also used within HTTP/1.1 to advise servers about which transfer + codings the client is able to accept in a response. + As of publication, only HTTP/1.1 uses transfer codings + (see Section 7 of [HTTP/1.1]).¶

+

+ The TE field value is a list of members, with each member (aside from + "trailers") consisting of a transfer coding name token with an optional + weight indicating the client's relative preference for that + transfer coding (Section 12.4.2) and + optional parameters for that transfer coding.¶

+ + + + +
+
  TE                 = #t-codings
+  t-codings          = "trailers" / ( transfer-coding [ weight ] )
+  transfer-coding    = token *( OWS ";" OWS transfer-parameter )
+  transfer-parameter = token BWS "=" BWS ( token / quoted-string )
+
¶ +
+

+ A sender of TE MUST also send a "TE" connection option within the + Connection header field (Section 7.6.1) + to inform intermediaries not to forward this field.¶

+
+
+
+
+

+10.1.5. User-Agent +

+ + + +

+ The "User-Agent" header field contains information about the user agent + originating the request, which is often used by servers to help identify + the scope of reported interoperability problems, to work around or tailor + responses to avoid particular user agent limitations, and for analytics + regarding browser or operating system use. A user agent SHOULD send + a User-Agent header field in each request unless specifically configured not + to do so.¶

+ +
+
  User-Agent = product *( RWS ( product / comment ) )
+
¶ +
+

+ The User-Agent field value consists of one or more product identifiers, + each followed by zero or more comments (Section 5.6.5), which together + identify the user agent software and its significant subproducts. + By convention, the product identifiers are listed in decreasing order of + their significance for identifying the user agent software. Each product + identifier consists of a name and optional version.¶

+ + +
+
  product         = token ["/" product-version]
+  product-version = token
+
¶ +
+

+ A sender SHOULD limit generated product identifiers to what is necessary + to identify the product; a sender MUST NOT generate advertising or other + nonessential information within the product identifier. + A sender SHOULD NOT generate information in product-version + that is not a version identifier (i.e., successive versions of the same + product name ought to differ only in the product-version portion of the + product identifier).¶

+

+ Example:¶

+
+
User-Agent: CERN-LineMode/2.15 libwww/2.17b3
+
¶ +
+

+ A user agent SHOULD NOT generate a User-Agent header field containing needlessly + fine-grained detail and SHOULD limit the addition of subproducts by third + parties. Overly long and detailed User-Agent field values increase request + latency and the risk of a user being identified against their wishes + ("fingerprinting").¶

+

+ Likewise, implementations are encouraged not to use the product tokens of + other implementations in order to declare compatibility with them, as this + circumvents the purpose of the field. If a user agent masquerades as a + different user agent, recipients can assume that the user intentionally + desires to see responses tailored for that identified user agent, even + if they might not work as well for the actual user agent being used.¶

+
+
+
+
+
+
+

+10.2. Response Context Fields +

+

+ The response header fields below provide additional information about the + response, beyond what is implied by the status code, including information + about the server, about the target resource, or about related + resources.¶

+
+
+

+10.2.1. Allow +

+ + + +

+ The "Allow" header field lists the set of methods advertised as + supported by the target resource. The purpose of this field + is strictly to inform the recipient of valid request methods associated + with the resource.¶

+ +
+
  Allow = #method
+
¶ +
+

+ Example of use:¶

+
+
Allow: GET, HEAD, PUT
+
¶ +
+

+ The actual set of allowed methods is defined by the origin server at the + time of each request. An origin server MUST generate an Allow header field in a + 405 (Method Not Allowed) response and MAY do so in any + other response. An empty Allow field value indicates that the resource + allows no methods, which might occur in a 405 response if the resource has + been temporarily disabled by configuration.¶

+

+ A proxy MUST NOT modify the Allow header field -- it does not need + to understand all of the indicated methods in order to handle them + according to the generic message handling rules.¶

+
+
+
+
+

+10.2.2. Location +

+ + + +

+ The "Location" header field is used in some responses to refer to a + specific resource in relation to the response. The type of relationship is + defined by the combination of request method and status code semantics.¶

+ +
+
  Location = URI-reference
+
¶ +
+

+ The field value consists of a single URI-reference. When it has the form + of a relative reference ([URI], Section 4.2), + the final value is computed by resolving it against the target + URI ([URI], Section 5).¶

+

+ For 201 (Created) responses, the Location value refers to + the primary resource created by the request. + For 3xx (Redirection) responses, the Location value refers + to the preferred target resource for automatically redirecting the request.¶

+

+ If the Location value provided in a 3xx (Redirection) + response does not have a fragment component, a user agent MUST process the + redirection as if the value inherits the fragment component of the URI + reference used to generate the target URI (i.e., the redirection + inherits the original reference's fragment, if any).¶

+

+ For example, a GET request generated for the URI reference + "http://www.example.org/~tim" might result in a + 303 (See Other) response containing the header field:¶

+
+
Location: /People.html#tim
+
¶ +
+

+ which suggests that the user agent redirect to + "http://www.example.org/People.html#tim"¶

+

+ Likewise, a GET request generated for the URI reference + "http://www.example.org/index.html#larry" might result in a + 301 (Moved Permanently) response containing the header + field:¶

+
+
Location: http://www.example.net/index.html
+
¶ +
+

+ which suggests that the user agent redirect to + "http://www.example.net/index.html#larry", preserving the original fragment + identifier.¶

+

+ There are circumstances in which a fragment identifier in a Location + value would not be appropriate. For example, the Location header field in a + 201 (Created) response is supposed to provide a URI that is + specific to the created resource.¶

+ + +
+
+
+
+

+10.2.3. Retry-After +

+ + + +

+ Servers send the "Retry-After" header field to indicate how long the user + agent ought to wait before making a follow-up request. When sent with a + 503 (Service Unavailable) response, Retry-After indicates + how long the service is expected to be unavailable to the client. + When sent with any 3xx (Redirection) response, Retry-After + indicates the minimum time that the user agent is asked to wait before + issuing the redirected request.¶

+

+ The Retry-After field value can be either an HTTP-date or a number + of seconds to delay after receiving the response.¶

+ +
+
  Retry-After = HTTP-date / delay-seconds
+
¶ +
+
+

+ + A delay-seconds value is a non-negative decimal integer, representing time + in seconds.¶

+
+ +
+
  delay-seconds  = 1*DIGIT
+
¶ +
+

+ Two examples of its use are¶

+
+
Retry-After: Fri, 31 Dec 1999 23:59:59 GMT
+Retry-After: 120
+
¶ +
+

+ In the latter example, the delay is 2 minutes.¶

+
+
+
+
+

+10.2.4. Server +

+ + + +

+ The "Server" header field contains information about the + software used by the origin server to handle the request, which is often + used by clients to help identify the scope of reported interoperability + problems, to work around or tailor requests to avoid particular server + limitations, and for analytics regarding server or operating system use. + An origin server MAY generate a Server header field in its responses.¶

+ +
+
  Server = product *( RWS ( product / comment ) )
+
¶ +
+

+ The Server header field value consists of one or more product identifiers, each + followed by zero or more comments (Section 5.6.5), which together + identify the origin server software and its significant subproducts. + By convention, the product identifiers are listed in decreasing order of + their significance for identifying the origin server software. Each product + identifier consists of a name and optional version, as defined in + Section 10.1.5.¶

+

+ Example:¶

+
+
Server: CERN/3.0 libwww/2.17
+
¶ +
+

+ An origin server SHOULD NOT generate a Server header field containing needlessly + fine-grained detail and SHOULD limit the addition of subproducts by third + parties. Overly long and detailed Server field values increase response + latency and potentially reveal internal implementation details that might + make it (slightly) easier for attackers to find and exploit known security + holes.¶

+
+
+
+
+
+
+
+
+

+11. HTTP Authentication +

+
+
+

+11.1. Authentication Scheme +

+

+ HTTP provides a general framework for access control and authentication, + via an extensible set of challenge-response authentication schemes, which + can be used by a server to challenge a client request and by a client to + provide authentication information. It uses a case-insensitive + token to identify the authentication scheme:¶

+ +
+
  auth-scheme    = token
+
¶ +
+

+ Aside from the general framework, this document does not specify any + authentication schemes. New and existing authentication schemes are + specified independently and ought to be registered within the + "Hypertext Transfer Protocol (HTTP) Authentication Scheme Registry". + For example, the "basic" and "digest" authentication schemes are defined by + [RFC7617] and + [RFC7616], respectively.¶

+
+
+
+
+

+11.2. Authentication Parameters +

+

+ The authentication scheme is followed by additional information necessary + for achieving authentication via that scheme as either a + comma-separated list of parameters or a single sequence of characters + capable of holding base64-encoded information.¶

+ +
+
  token68        = 1*( ALPHA / DIGIT /
+                       "-" / "." / "_" / "~" / "+" / "/" ) *"="
+
¶ +
+

+ The token68 syntax allows the 66 unreserved URI characters + ([URI]), plus a few others, so that it can hold a + base64, base64url (URL and filename safe alphabet), base32, or base16 (hex) + encoding, with or without padding, but excluding whitespace + ([RFC4648]).¶

+

+ Authentication parameters are name/value pairs, where the name token is + matched case-insensitively + and each parameter name MUST only occur once per challenge.¶

+ +
+
  auth-param     = token BWS "=" BWS ( token / quoted-string )
+
¶ +
+

+ Parameter values can be expressed either as "token" or as "quoted-string" + (Section 5.6). + Authentication scheme definitions need to accept both notations, both for + senders and recipients, to allow recipients to use generic parsing + components regardless of the authentication scheme.¶

+

+ For backwards compatibility, authentication scheme definitions can restrict + the format for senders to one of the two variants. This can be important + when it is known that deployed implementations will fail when encountering + one of the two formats.¶

+
+
+
+
+

+11.3. Challenge and Response +

+

+ A 401 (Unauthorized) response message is used by an origin + server to challenge the authorization of a user agent, including a + WWW-Authenticate header field containing at least one + challenge applicable to the requested resource.¶

+

+ A 407 (Proxy Authentication Required) response message is + used by a proxy to challenge the authorization of a client, including a + Proxy-Authenticate header field containing at least one + challenge applicable to the proxy for the requested resource.¶

+ +
+
  challenge   = auth-scheme [ 1*SP ( token68 / #auth-param ) ]
+
¶ +
+ +

+ A user agent that wishes to authenticate itself with an origin server + -- usually, but not necessarily, after receiving a + 401 (Unauthorized) -- can do so by including an + Authorization header field with the request.¶

+

+ A client that wishes to authenticate itself with a proxy -- usually, + but not necessarily, after receiving a + 407 (Proxy Authentication Required) -- can do so by + including a Proxy-Authorization header field with the + request.¶

+
+
+
+
+

+11.4. Credentials +

+

+ Both the Authorization field value and the + Proxy-Authorization field value contain the client's + credentials for the realm of the resource being requested, based upon a + challenge received in a response (possibly at some point in the past). + When creating their values, the user agent ought to do so by selecting the + challenge with what it considers to be the most secure auth-scheme that it + understands, obtaining credentials from the user as appropriate. + Transmission of credentials within header field values implies significant + security considerations regarding the confidentiality of the underlying + connection, as described in + Section 17.16.1.¶

+ +
+
  credentials = auth-scheme [ 1*SP ( token68 / #auth-param ) ]
+
¶ +
+

+ Upon receipt of a request for a protected resource that omits credentials, + contains invalid credentials (e.g., a bad password) or partial credentials + (e.g., when the authentication scheme requires more than one round trip), + an origin server SHOULD send a 401 (Unauthorized) response + that contains a WWW-Authenticate header field with at least + one (possibly new) challenge applicable to the requested resource.¶

+

+ Likewise, upon receipt of a request that omits proxy credentials or + contains invalid or partial proxy credentials, a proxy that requires + authentication SHOULD generate a + 407 (Proxy Authentication Required) response that contains + a Proxy-Authenticate header field with at least one + (possibly new) challenge applicable to the proxy.¶

+

+ A server that receives valid credentials that are not adequate to gain + access ought to respond with the 403 (Forbidden) status + code (Section 15.5.4).¶

+

+ HTTP does not restrict applications to this simple challenge-response + framework for access authentication. Additional mechanisms can be used, + such as authentication at the transport level or via message encapsulation, + and with additional header fields specifying authentication information. + However, such additional mechanisms are not defined by this specification.¶

+

+ Note that various custom mechanisms for user authentication use the + Set-Cookie and Cookie header fields, defined in [COOKIE], + for passing tokens related to authentication.¶

+
+
+
+
+

+11.5. Establishing a Protection Space (Realm) +

+ + + +

+ The "realm" authentication parameter is reserved for use by + authentication schemes that wish to indicate a scope of protection.¶

+

+ A "protection space" is defined by the origin (see + Section 4.3.1) of the + server being accessed, in combination with the realm value if present. + These realms allow the protected resources on a server to be + partitioned into a set of protection spaces, each with its own + authentication scheme and/or authorization database. The realm value + is a string, generally assigned by the origin server, that can have + additional semantics specific to the authentication scheme. Note that a + response can have multiple challenges with the same auth-scheme but + with different realms.¶

+

+ The protection space determines the domain over which credentials can + be automatically applied. If a prior request has been authorized, the + user agent MAY reuse the same credentials for all other requests within + that protection space for a period of time determined by the authentication + scheme, parameters, and/or user preferences (such as a configurable + inactivity timeout).¶

+

+ The extent of a protection space, and therefore the requests to which + credentials might be automatically applied, is not necessarily known to + clients without additional information. An authentication scheme might + define parameters that describe the extent of a protection space. Unless + specifically allowed by the authentication scheme, a single protection + space cannot extend outside the scope of its server.¶

+

+ For historical reasons, a sender MUST only generate the quoted-string syntax. + Recipients might have to support both token and quoted-string syntax for + maximum interoperability with existing clients that have been accepting both + notations for a long time.¶

+
+
+
+
+

+11.6. Authenticating Users to Origin Servers +

+
+
+

+11.6.1. WWW-Authenticate +

+ + + +

+ The "WWW-Authenticate" response header field indicates the authentication + scheme(s) and parameters applicable to the target resource.¶

+ +
+
  WWW-Authenticate = #challenge
+
¶ +
+

+ A server generating a 401 (Unauthorized) response + MUST send a WWW-Authenticate header field containing at least one + challenge. A server MAY generate a WWW-Authenticate header field + in other response messages to indicate that supplying credentials + (or different credentials) might affect the response.¶

+

+ A proxy forwarding a response MUST NOT modify any + WWW-Authenticate header fields in that response.¶

+

+ User agents are advised to take special care in parsing the field value, as + it might contain more than one challenge, and each challenge can contain a + comma-separated list of authentication parameters. Furthermore, the header + field itself can occur multiple times.¶

+

+ For instance:¶

+
+
WWW-Authenticate: Basic realm="simple", Newauth realm="apps",
+                 type=1, title="Login to \"apps\""
+
¶ +
+

+ This header field contains two challenges, one for the "Basic" scheme with + a realm value of "simple" and another for the "Newauth" scheme with a + realm value of "apps". It also contains two additional parameters, "type" and "title".¶

+

+ Some user agents do not recognize this form, however. As a result, sending + a WWW-Authenticate field value with more than one member on the same field + line might not be interoperable.¶

+ +
+
+
+
+

+11.6.2. Authorization +

+ + + +

+ The "Authorization" header field allows a user agent to authenticate itself + with an origin server -- usually, but not necessarily, after receiving + a 401 (Unauthorized) response. Its value consists of + credentials containing the authentication information of the user agent for + the realm of the resource being requested.¶

+ +
+
  Authorization = credentials
+
¶ +
+

+ If a request is authenticated and a realm specified, the same credentials + are presumed to be valid for all other requests within this realm (assuming + that the authentication scheme itself does not require otherwise, such as + credentials that vary according to a challenge value or using synchronized + clocks).¶

+

+ A proxy forwarding a request MUST NOT modify any + Authorization header fields in that request. + See Section 3.5 of [CACHING] for details of and requirements + pertaining to handling of the Authorization header field by HTTP caches.¶

+
+
+
+
+

+11.6.3. Authentication-Info +

+ + + +

+ HTTP authentication schemes can use the "Authentication-Info" response + field to communicate information after the client's authentication credentials have been accepted. + This information can include a finalization message from the server (e.g., it can contain the + server authentication).¶

+

+ The field value is a list of parameters (name/value pairs), using the "auth-param" + syntax defined in Section 11.3. + This specification only describes the generic format; authentication schemes + using Authentication-Info will define the individual parameters. The "Digest" + Authentication Scheme, for instance, defines multiple parameters in + Section 3.5 of [RFC7616].¶

+ +
+
  Authentication-Info = #auth-param
+
¶ +
+

+ The Authentication-Info field can be used in any HTTP response, + independently of request method and status code. Its semantics are defined + by the authentication scheme indicated by the Authorization header field + (Section 11.6.2) of the corresponding request.¶

+

+ A proxy forwarding a response is not allowed to modify the field value in any + way.¶

+

+ Authentication-Info can be sent as a trailer field + (Section 6.5) + when the authentication scheme explicitly allows this.¶

+
+
+
+
+
+
+

+11.7. Authenticating Clients to Proxies +

+
+
+

+11.7.1. Proxy-Authenticate +

+ + + +

+ The "Proxy-Authenticate" header field consists of at least one + challenge that indicates the authentication scheme(s) and parameters + applicable to the proxy for this request. + A proxy MUST send at least one Proxy-Authenticate header field in + each 407 (Proxy Authentication Required) response that it + generates.¶

+ +
+
  Proxy-Authenticate = #challenge
+
¶ +
+

+ Unlike WWW-Authenticate, the Proxy-Authenticate header field + applies only to the next outbound client on the response chain. + This is because only the client that chose a given proxy is likely to have + the credentials necessary for authentication. However, when multiple + proxies are used within the same administrative domain, such as office and + regional caching proxies within a large corporate network, it is common + for credentials to be generated by the user agent and passed through the + hierarchy until consumed. Hence, in such a configuration, it will appear + as if Proxy-Authenticate is being forwarded because each proxy will send + the same challenge set.¶

+

+ Note that the parsing considerations for WWW-Authenticate + apply to this header field as well; see Section 11.6.1 + for details.¶

+
+
+
+
+

+11.7.2. Proxy-Authorization +

+ + + +

+ The "Proxy-Authorization" header field allows the client to + identify itself (or its user) to a proxy that requires + authentication. Its value consists of credentials containing the + authentication information of the client for the proxy and/or realm of the + resource being requested.¶

+ +
+
  Proxy-Authorization = credentials
+
¶ +
+

+ Unlike Authorization, the Proxy-Authorization header field + applies only to the next inbound proxy that demanded authentication using + the Proxy-Authenticate header field. When multiple proxies are used + in a chain, the Proxy-Authorization header field is consumed by the first + inbound proxy that was expecting to receive credentials. A proxy MAY + relay the credentials from the client request to the next proxy if that is + the mechanism by which the proxies cooperatively authenticate a given + request.¶

+
+
+
+
+

+11.7.3. Proxy-Authentication-Info +

+ + + +

+ The "Proxy-Authentication-Info" response header field is equivalent to + Authentication-Info, except that it applies to proxy authentication (Section 11.3) + and its semantics are defined by the + authentication scheme indicated by the Proxy-Authorization header field + (Section 11.7.2) + of the corresponding request:¶

+ +
+
  Proxy-Authentication-Info = #auth-param
+
¶ +
+

+ However, unlike Authentication-Info, the Proxy-Authentication-Info header + field applies only to the next outbound client on the response chain. This is + because only the client that chose a given proxy is likely to have the + credentials necessary for authentication. However, when multiple proxies are + used within the same administrative domain, such as office and regional + caching proxies within a large corporate network, it is common for + credentials to be generated by the user agent and passed through the + hierarchy until consumed. Hence, in such a configuration, it will appear as + if Proxy-Authentication-Info is being forwarded because each proxy will send + the same field value.¶

+

+ Proxy-Authentication-Info can be sent as a trailer field + (Section 6.5) + when the authentication scheme explicitly allows this.¶

+
+
+
+
+
+
+
+
+

+12. Content Negotiation +

+

+ When responses convey content, whether indicating a success or + an error, the origin server often has different ways of representing that + information; for example, in different formats, languages, or encodings. + Likewise, different users or user agents might have differing capabilities, + characteristics, or preferences that could influence which representation, + among those available, would be best to deliver. For this reason, HTTP + provides mechanisms for content negotiation.¶

+

+ This specification defines three patterns of content negotiation that can + be made visible within the protocol: + "proactive" negotiation, where the server selects the representation based + upon the user agent's stated preferences; "reactive" negotiation, + where the server provides a list of representations for the user agent to + choose from; and "request content" negotiation, where the user agent + selects the representation for a future request based upon the server's + stated preferences in past responses.¶

+

+ Other patterns of content negotiation include + "conditional content", where the representation consists of multiple + parts that are selectively rendered based on user agent parameters, + "active content", where the representation contains a script that + makes additional (more specific) requests based on the user agent + characteristics, and "Transparent Content Negotiation" + ([RFC2295]), where content selection is performed by + an intermediary. These patterns are not mutually exclusive, and each has + trade-offs in applicability and practicality.¶

+

+ Note that, in all cases, HTTP is not aware of the resource semantics. + The consistency with which an origin server responds to requests, over time + and over the varying dimensions of content negotiation, and thus the + "sameness" of a resource's observed representations over time, is + determined entirely by whatever entity or algorithm selects or generates + those responses.¶

+
+
+

+12.1. Proactive Negotiation +

+

+ When content negotiation preferences are sent by the user agent in a + request to encourage an algorithm located at the server to + select the preferred representation, it is called + "proactive negotiation" + (a.k.a., "server-driven negotiation"). Selection is based on + the available representations for a response (the dimensions over which it + might vary, such as language, content coding, etc.) compared to various + information supplied in the request, including both the explicit + negotiation header fields below and implicit + characteristics, such as the client's network address or parts of the + User-Agent field.¶

+

+ Proactive negotiation is advantageous when the algorithm for + selecting from among the available representations is difficult to + describe to a user agent, or when the server desires to send its + "best guess" to the user agent along with the first response (when that + "best guess" is good enough for the user, this avoids the round-trip + delay of a subsequent request). In order to improve the server's + guess, a user agent MAY send request header fields that describe + its preferences.¶

+

+ Proactive negotiation has serious disadvantages:¶

+
    +
  • + It is impossible for the server to accurately determine what + might be "best" for any given user, since that would require + complete knowledge of both the capabilities of the user agent + and the intended use for the response (e.g., does the user want + to view it on screen or print it on paper?);¶ +
    • +
  • + Having the user agent describe its capabilities in every + request can be both very inefficient (given that only a small + percentage of responses have multiple representations) and a + potential risk to the user's privacy;¶ +
    • +
  • + It complicates the implementation of an origin server and the + algorithms for generating responses to a request; and,¶ +
    • +
  • + It limits the reusability of responses for shared caching.¶ +
    • +
+

+ A user agent cannot rely on proactive negotiation preferences being + consistently honored, since the origin server might not implement proactive + negotiation for the requested resource or might decide that sending a + response that doesn't conform to the user agent's preferences is better + than sending a 406 (Not Acceptable) response.¶

+

+ A Vary header field (Section 12.5.5) is + often sent in a response subject to proactive negotiation to indicate what + parts of the request information were used in the selection algorithm.¶

+

+ The request header fields Accept, + Accept-Charset, Accept-Encoding, and + Accept-Language are defined below for a user agent to engage + in proactive negotiation of the response content. + The preferences sent in these + fields apply to any content in the response, including representations of + the target resource, representations of error or processing status, and + potentially even the miscellaneous text strings that might appear within + the protocol.¶

+
+
+
+
+

+12.2. Reactive Negotiation +

+

+ With "reactive negotiation" (a.k.a., "agent-driven negotiation"), selection of + content (regardless of the status code) is performed by + the user agent after receiving an initial response. The mechanism for + reactive negotiation might be as simple as a list of references to + alternative representations.¶

+

+ If the user agent is not satisfied by the initial response content, + it can perform a GET request on one or more of the alternative resources + to obtain a different representation. Selection of such alternatives might + be performed automatically (by the user agent) or manually (e.g., by the + user selecting from a hypertext menu).¶

+

+ A server might choose not to send an initial representation, other than + the list of alternatives, and thereby indicate that reactive + negotiation by the user agent is preferred. For example, the alternatives + listed in responses with the 300 (Multiple Choices) and + 406 (Not Acceptable) status codes include information about + available representations so that the user or user agent can react by + making a selection.¶

+

+ Reactive negotiation is advantageous when the response would vary + over commonly used dimensions (such as type, language, or encoding), + when the origin server is unable to determine a user agent's + capabilities from examining the request, and generally when public + caches are used to distribute server load and reduce network usage.¶

+

+ Reactive negotiation suffers from the disadvantages of transmitting + a list of alternatives to the user agent, which degrades user-perceived + latency if transmitted in the header section, and needing a second request + to obtain an alternate representation. Furthermore, this specification + does not define a mechanism for supporting automatic selection, though it + does not prevent such a mechanism from being developed.¶

+
+
+
+
+

+12.3. Request Content Negotiation +

+

+ When content negotiation preferences are sent in a server's response, the + listed preferences are called "request content negotiation" + because they intend to influence selection of an appropriate content for + subsequent requests to that resource. For example, + the Accept (Section 12.5.1) and + Accept-Encoding (Section 12.5.3) + header fields can be sent in a response to indicate preferred media types + and content codings for subsequent requests to that resource.¶

+

+ Similarly, Section 3.1 of [RFC5789] defines + the "Accept-Patch" response header field, which allows discovery of + which content types are accepted in PATCH requests.¶

+
+
+
+
+

+12.4. Content Negotiation Field Features +

+
+
+

+12.4.1. Absence +

+

+ For each of the content negotiation fields, a request that does not contain + the field implies that the sender has no preference on that dimension of + negotiation.¶

+

+ If a content negotiation header field is present in a request and none of + the available + representations for the response can be considered acceptable according to + it, the origin server can either honor the header field by sending a + 406 (Not Acceptable) response or disregard the header field + by treating the response as if it is not subject to content negotiation + for that request header field. This does not imply, however, that the + client will be able to use the representation.¶

+ +
+
+
+
+

+12.4.2. Quality Values +

+

+ The content negotiation fields defined by this specification + use a common parameter, named "q" (case-insensitive), to assign a relative + "weight" to the preference for that associated kind of content. + This weight is referred to as a "quality value" (or "qvalue") because + the same parameter name is often used within server configurations to + assign a weight to the relative quality of the various representations + that can be selected for a resource.¶

+

+ The weight is normalized to a real number in the range 0 through 1, + where 0.001 is the least preferred and 1 is the most preferred; + a value of 0 means "not acceptable". If no "q" parameter is present, + the default weight is 1.¶

+ + +
+
  weight = OWS ";" OWS "q=" qvalue
+  qvalue = ( "0" [ "." 0*3DIGIT ] )
+         / ( "1" [ "." 0*3("0") ] )
+
¶ +
+

+ A sender of qvalue MUST NOT generate more than three digits after the + decimal point. User configuration of these values ought to be limited in + the same fashion.¶

+
+
+
+
+

+12.4.3. Wildcard Values +

+

+ Most of these header fields, where indicated, define a wildcard value ("*") + to select unspecified values. If no wildcard is present, values that are + not explicitly mentioned in the field are considered unacceptable. + Within Vary, the wildcard value means that the variance + is unlimited.¶

+ +
+
+
+
+
+
+

+12.5. Content Negotiation Fields +

+
+
+

+12.5.1. Accept +

+ + + +

+ The "Accept" header field can be used by user agents to specify their + preferences regarding response media types. For example, Accept header + fields can be used to indicate that the request is specifically limited to + a small set of desired types, as in the case of a request for an in-line + image.¶

+

+ When sent by a server in a response, Accept provides information + about which content types are preferred in the content of a subsequent + request to the same resource.¶

+ + +
+
  Accept = #( media-range [ weight ] )
+
+  media-range    = ( "*/*"
+                     / ( type "/" "*" )
+                     / ( type "/" subtype )
+                   ) parameters
+
¶ +
+

+ The asterisk "*" character is used to group media types into ranges, + with "*/*" indicating all media types and "type/*" indicating all + subtypes of that type. The media-range can include media type + parameters that are applicable to that range.¶

+

+ Each media-range might be followed by optional applicable media type + parameters (e.g., charset), followed by an optional "q" + parameter for indicating a relative weight (Section 12.4.2).¶

+

+ Previous specifications allowed additional extension parameters to appear + after the weight parameter. The accept extension grammar (accept-params, accept-ext) has + been removed because it had a complicated definition, was not being used in + practice, and is more easily deployed through new header fields. Senders + using weights SHOULD send "q" last (after all media-range parameters). + Recipients SHOULD process any parameter named "q" as weight, regardless of + parameter ordering.¶

+ +

+ The example¶

+
+
Accept: audio/*; q=0.2, audio/basic
+
¶ +
+

+ is interpreted as "I prefer audio/basic, but send me any audio + type if it is the best available after an 80% markdown in quality".¶

+

+ A more elaborate example is¶

+
+
Accept: text/plain; q=0.5, text/html,
+       text/x-dvi; q=0.8, text/x-c
+
¶ +
+

+ Verbally, this would be interpreted as "text/html and text/x-c are + the equally preferred media types, but if they do not exist, then send the + text/x-dvi representation, and if that does not exist, send the text/plain + representation".¶

+

+ Media ranges can be overridden by more specific media ranges or + specific media types. If more than one media range applies to a given + type, the most specific reference has precedence. For example,¶

+
+
Accept: text/*, text/plain, text/plain;format=flowed, */*
+
¶ +
+

+ have the following precedence:¶

+
    +
  1. text/plain;format=flowed¶ +
    2. +
  2. text/plain¶ +
    3. +
  3. text/*¶ +
    4. +
  4. */*¶ +
    5. +
+

+ The media type quality factor associated with a given type is + determined by finding the media range with the highest precedence + that matches the type. For example,¶

+
+
Accept: text/*;q=0.3, text/plain;q=0.7, text/plain;format=flowed,
+       text/plain;format=fixed;q=0.4, */*;q=0.5
+
¶ +
+

+ would cause the following values to be associated:¶

+ + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +
Table 5
Media TypeQuality Value
text/plain;format=flowed1
text/plain0.7
text/html0.3
image/jpeg0.5
text/plain;format=fixed0.4
text/html;level=30.7
+ +
+
+
+
+

+12.5.2. Accept-Charset +

+ + + +

+ The "Accept-Charset" header field can be sent by a user agent to indicate + its preferences for charsets in textual response content. For example, + this field allows user agents capable of understanding more comprehensive + or special-purpose charsets to signal that capability to an origin server + that is capable of representing information in those charsets.¶

+ +
+
  Accept-Charset = #( ( token / "*" ) [ weight ] )
+
¶ +
+

+ Charset names are defined in Section 8.3.2. + A user agent MAY associate a quality value with each charset to indicate + the user's relative preference for that charset, as defined in Section 12.4.2. + An example is¶

+
+
Accept-Charset: iso-8859-5, unicode-1-1;q=0.8
+
¶ +
+

+ The special value "*", if present in the Accept-Charset header field, + matches every charset that is not mentioned elsewhere in the + field.¶

+ +
+
+
+
+

+12.5.3. Accept-Encoding +

+ + + +

+ The "Accept-Encoding" header field can be used to indicate preferences + regarding the use of content codings (Section 8.4.1).¶

+

+ When sent by a user agent in a request, Accept-Encoding indicates the + content codings acceptable in a response.¶

+

+ When sent by a server in a response, Accept-Encoding provides information + about which content codings are preferred in the content of a subsequent + request to the same resource.¶

+

+ An "identity" token is used as a synonym for + "no encoding" in order to communicate when no encoding is preferred.¶

+ + +
+
  Accept-Encoding  = #( codings [ weight ] )
+  codings          = content-coding / "identity" / "*"
+
¶ +
+

+ Each codings value MAY be given an associated quality value (weight) + representing the preference for that encoding, as defined in Section 12.4.2. + The asterisk "*" symbol in an Accept-Encoding field matches any available + content coding not explicitly listed in the field.¶

+

+ Examples:¶

+
+
Accept-Encoding: compress, gzip
+Accept-Encoding:
+Accept-Encoding: *
+Accept-Encoding: compress;q=0.5, gzip;q=1.0
+Accept-Encoding: gzip;q=1.0, identity; q=0.5, *;q=0
+
¶ +
+

+ A server tests whether a content coding for a given representation is + acceptable using these rules:¶

+
    +
  1. If no Accept-Encoding header field is in the request, any content coding is + considered acceptable by the user agent.¶ +
    2. +
  2. If the representation has no content coding, then it is acceptable + by default unless specifically excluded by the Accept-Encoding header field + stating either "identity;q=0" or "*;q=0" without a more specific + entry for "identity".¶ +
    3. +
  3. If the representation's content coding is one of the content codings + listed in the Accept-Encoding field value, then it is acceptable unless + it is accompanied by a qvalue of 0. (As defined in Section 12.4.2, a + qvalue of 0 means "not acceptable".)¶ +
    4. +
+

+ A representation could be encoded with multiple content codings. However, most + content codings are alternative ways to accomplish the same purpose + (e.g., data compression). When selecting between multiple content codings that + have the same purpose, the acceptable content coding with the highest + non-zero qvalue is preferred.¶

+

+ An Accept-Encoding header field with a field value that is empty + implies that the user agent does not want any content coding in response. + If a non-empty Accept-Encoding header field is present in a request and none of the + available representations for the response have a content coding that + is listed as acceptable, the origin server SHOULD send a response + without any content coding unless the identity coding is indicated as unacceptable.¶

+

+ When the Accept-Encoding header field is present in a response, it indicates + what content codings the resource was willing to accept in the associated + request. The field value is evaluated the same way as in a request.¶

+

+ Note that this information is specific to the associated request; the set of + supported encodings might be different for other resources on the same + server and could change over time or depend on other aspects of the request + (such as the request method).¶

+

+ Servers that fail a request due to an unsupported content coding ought to + respond with a 415 (Unsupported Media Type) status and + include an Accept-Encoding header field in that response, allowing + clients to distinguish between issues related to content codings and media + types. In order to avoid confusion with issues related to media types, + servers that fail a request with a 415 status for reasons unrelated to + content codings MUST NOT include the Accept-Encoding header + field.¶

+

+ The most common use of Accept-Encoding is in responses with a + 415 (Unsupported Media Type) status code, in response to + optimistic use of a content coding by clients. However, the header field + can also be used to indicate to clients that content codings are supported in order + to optimize future interactions. For example, a resource might include it + in a 2xx (Successful) response when the request content was + big enough to justify use of a compression coding but the client failed do + so.¶

+
+
+
+
+

+12.5.4. Accept-Language +

+ + + +

+ The "Accept-Language" header field can be used by user agents to + indicate the set of natural languages that are preferred in the response. + Language tags are defined in Section 8.5.1.¶

+ + +
+
  Accept-Language = #( language-range [ weight ] )
+  language-range  =
+            <language-range, see [RFC4647], Section 2.1>
+
¶ +
+

+ Each language-range can be given an associated quality value + representing an estimate of the user's preference for the languages + specified by that range, as defined in Section 12.4.2. For example,¶

+
+
Accept-Language: da, en-gb;q=0.8, en;q=0.7
+
¶ +
+

+ would mean: "I prefer Danish, but will accept British English and + other types of English".¶

+

+ Note that some recipients treat the order in which language tags are listed + as an indication of descending priority, particularly for tags that are + assigned equal quality values (no value is the same as q=1). However, this + behavior cannot be relied upon. For consistency and to maximize + interoperability, many user agents assign each language tag a unique + quality value while also listing them in order of decreasing quality. + Additional discussion of language priority lists can be found in + Section 2.3 of [RFC4647].¶

+

+ For matching, Section 3 of [RFC4647] defines + several matching schemes. Implementations can offer the most appropriate + matching scheme for their requirements. The "Basic Filtering" scheme + ([RFC4647], Section 3.3.1) is identical to the + matching scheme that was previously defined for HTTP in + Section 14.4 of [RFC2616].¶

+

+ It might be contrary to the privacy expectations of the user to send + an Accept-Language header field with the complete linguistic preferences of + the user in every request (Section 17.13).¶

+

+ Since intelligibility is highly dependent on the individual user, user + agents need to allow user control over the linguistic preference (either + through configuration of the user agent itself or by defaulting to a user + controllable system setting). + A user agent that does not provide such control to the user MUST NOT + send an Accept-Language header field.¶

+ +
+
+
+
+

+12.5.5. Vary +

+ + + +

+ The "Vary" header field in a response describes what parts of a request + message, aside from the method and target URI, might have influenced the + origin server's process for selecting the content of this response.¶

+ +
+
  Vary = #( "*" / field-name )
+
¶ +
+

+ A Vary field value is either the wildcard member "*" or a list of + request field names, known as the selecting header fields, that might + have had a role in selecting the representation for this response. + Potential selecting header fields are not limited to fields defined by + this specification.¶

+

+ A list containing the member "*" signals that other aspects of the + request might have played a role in selecting the response representation, + possibly including aspects outside the message syntax (e.g., the + client's network address). + A recipient will not be able to determine whether this response is + appropriate for a later request without forwarding the request to the + origin server. A proxy MUST NOT generate "*" in a Vary field value.¶

+

+ For example, a response that contains¶

+
+
Vary: accept-encoding, accept-language
+
¶ +
+

+ indicates that the origin server might have used the request's + Accept-Encoding and Accept-Language + header fields (or lack thereof) as determining factors while choosing + the content for this response.¶

+

+ A Vary field containing a list of field names has two purposes:¶

+
    +
  1. +

    + To inform cache recipients that they MUST NOT use this response + to satisfy a later request unless the later request has the + same values for the listed header fields as the original request + (Section 4.1 of [CACHING]) or reuse of the + response has been validated by the origin server. + In other words, Vary expands the cache key + required to match a new request to the stored cache entry.¶

    +
    2. +
  2. +

    + To inform user agent recipients that this response was subject to + content negotiation (Section 12) and a + different representation might be sent in a subsequent request if + other values are provided in the listed header fields + (proactive negotiation).¶

    +
    3. +
+

+ An origin server SHOULD generate a Vary header field on a cacheable + response when it wishes that response to be selectively reused for + subsequent requests. Generally, that is the case when the response + content has been tailored to better fit the preferences expressed by + those selecting header fields, such as when an origin server has + selected the response's language based on the request's + Accept-Language header field.¶

+

+ Vary might be elided when an origin server considers variance in + content selection to be less significant than Vary's performance impact + on caching, particularly when reuse is already limited by cache + response directives (Section 5.2 of [CACHING]).¶

+

+ There is no need to send the Authorization field name in Vary because + reuse of that response for a different user is prohibited by the field + definition (Section 11.6.2). + Likewise, if the response content has been selected or influenced by + network region, but the origin server wants the cached response to be + reused even if recipients move from one region to another, then there + is no need for the origin server to indicate such variance in Vary.¶

+
+
+
+
+
+
+
+
+

+13. Conditional Requests +

+ +

+ A conditional request is an HTTP request with one or more request header + fields that indicate a precondition to be tested before + applying the request method to the target resource. + Section 13.2 defines when to evaluate preconditions and + their order of precedence when more than one precondition is present.¶

+

+ Conditional GET requests are the most efficient mechanism for HTTP + cache updates [CACHING]. Conditionals can also be + applied to state-changing methods, such as PUT and DELETE, to prevent + the "lost update" problem: one client accidentally overwriting + the work of another client that has been acting in parallel.¶

+
+
+

+13.1. Preconditions +

+ +

+ Preconditions are usually defined with respect to a state of the target + resource as a whole (its current value set) or the state as observed in a + previously obtained representation (one value in that set). If a resource + has multiple current representations, each with its own observable state, + a precondition will assume that the mapping of each request to a + selected representation (Section 3.2) + is consistent over time. + Regardless, if the mapping is inconsistent or the server is unable to + select an appropriate representation, then no harm will result when the + precondition evaluates to false.¶

+

+ Each precondition defined below consists of a comparison between a + set of validators obtained from prior representations of the target + resource to the current state of validators for the selected + representation (Section 8.8). Hence, these + preconditions evaluate whether the state of the target resource has + changed since a given state known by the client. The effect of such an + evaluation depends on the method semantics and choice of conditional, as + defined in Section 13.2.¶

+

+ Other preconditions, defined by other specifications as extension fields, + might place conditions on all recipients, on the state of the target + resource in general, or on a group of resources. For instance, the "If" + header field in WebDAV can make a request conditional on various aspects + of multiple resources, such as locks, if the recipient understands and + implements that field ([WEBDAV], Section 10.4).¶

+

+ Extensibility of preconditions is only possible when the precondition can + be safely ignored if unknown (like If-Modified-Since), when + deployment can be assumed for a given use case, or when implementation + is signaled by some other property of the target resource. This encourages + a focus on mutually agreed deployment of common standards.¶

+
+
+

+13.1.1. If-Match +

+ + + +

+ The "If-Match" header field makes the request method conditional on the + recipient origin server either having at least one current + representation of the target resource, when the field value is "*", or + having a current representation of the target resource that has an + entity tag matching a member of the list of entity tags provided in the + field value.¶

+

+ An origin server MUST use the strong comparison function when comparing + entity tags for If-Match (Section 8.8.3.2), since + the client intends this precondition to prevent the method from being + applied if there have been any changes to the representation data.¶

+ +
+
  If-Match = "*" / #entity-tag
+
¶ +
+

+ Examples:¶

+
+
If-Match: "xyzzy"
+If-Match: "xyzzy", "r2d2xxxx", "c3piozzzz"
+If-Match: *
+
¶ +
+

+ If-Match is most often used with state-changing methods (e.g., POST, PUT, + DELETE) to prevent accidental overwrites when multiple user agents might be + acting in parallel on the same resource (i.e., to prevent the "lost update" + problem). In general, it can be used with any method that involves the + selection or modification of a representation to abort the request if the + selected representation's current entity tag is not a + member within the If-Match field value.¶

+

+ When an origin server receives a request that selects a representation + and that request includes an If-Match header field, + the origin server MUST evaluate the If-Match condition per + Section 13.2 prior to performing the method.¶

+

+ To evaluate a received If-Match header field:¶

+
    +
  1. + If the field value is "*", the condition is true if the origin server + has a current representation for the target resource.¶ +
    2. +
  2. + If the field value is a list of entity tags, the condition is true if + any of the listed tags match the entity tag of the selected representation.¶ +
    3. +
  3. + Otherwise, the condition is false.¶ +
    4. +
+

+ An origin server that evaluates an If-Match condition MUST NOT perform + the requested method if the condition evaluates to false. Instead, + the origin server MAY + indicate that the conditional request failed by responding with a + 412 (Precondition Failed) status code. Alternatively, + if the request is a state-changing operation that appears to have already + been applied to the selected representation, the origin server MAY respond + with a 2xx (Successful) status code + (i.e., the change requested by the user agent has already succeeded, but + the user agent might not be aware of it, perhaps because the prior response + was lost or an equivalent change was made by some other user agent).¶

+

+ Allowing an origin server to send a success response when a change request + appears to have already been applied is more efficient for many authoring + use cases, but comes with some risk if multiple user agents are making + change requests that are very similar but not cooperative. + For example, multiple user agents writing to a common resource as a + semaphore (e.g., a nonatomic increment) are likely to collide and + potentially lose important state transitions. For those kinds of resources, + an origin server is better off being stringent in sending 412 for every + failed precondition on an unsafe method. + In other cases, excluding the ETag field from a success response might + encourage the user agent to perform a GET as its next request to eliminate + confusion about the resource's current state.¶

+

+ A client MAY send an If-Match header field in a + GET request to indicate that it would prefer a + 412 (Precondition Failed) response if the selected + representation does not match. However, this is only useful in range + requests (Section 14) for completing a previously + received partial representation when there is no desire for a new + representation. If-Range (Section 13.1.5) + is better suited for range requests when the client prefers to receive a + new representation.¶

+

+ A cache or intermediary MAY ignore If-Match because its + interoperability features are only necessary for an origin server.¶

+

+ Note that an If-Match header field with a list value containing "*" and + other values (including other instances of "*") is syntactically + invalid (therefore not allowed to be generated) and furthermore is + unlikely to be interoperable.¶

+
+
+
+
+

+13.1.2. If-None-Match +

+ + + +

+ The "If-None-Match" header field makes the request method conditional on + a recipient cache or origin server either not having any current + representation of the target resource, when the field value is "*", or + having a selected representation with an entity tag that does not match any + of those listed in the field value.¶

+

+ A recipient MUST use the weak comparison function when comparing + entity tags for If-None-Match (Section 8.8.3.2), + since weak entity tags can be used for cache validation even if there have + been changes to the representation data.¶

+ +
+
  If-None-Match = "*" / #entity-tag
+
¶ +
+

+ Examples:¶

+
+
If-None-Match: "xyzzy"
+If-None-Match: W/"xyzzy"
+If-None-Match: "xyzzy", "r2d2xxxx", "c3piozzzz"
+If-None-Match: W/"xyzzy", W/"r2d2xxxx", W/"c3piozzzz"
+If-None-Match: *
+
¶ +
+

+ If-None-Match is primarily used in conditional GET requests to enable + efficient updates of cached information with a minimum amount of + transaction overhead. When a client desires to update one or more stored + responses that have entity tags, the client SHOULD generate an + If-None-Match header field containing a list of those entity tags when + making a GET request; this allows recipient servers to send a + 304 (Not Modified) response to indicate when one of those + stored responses matches the selected representation.¶

+

+ If-None-Match can also be used with a value of "*" to prevent an unsafe + request method (e.g., PUT) from inadvertently modifying an existing + representation of the target resource when the client believes that + the resource does not have a current representation (Section 9.2.1). + This is a variation on the "lost update" problem that might arise if more + than one client attempts to create an initial representation for the target + resource.¶

+

+ When an origin server receives a request that selects a representation + and that request includes an If-None-Match header field, + the origin server MUST evaluate the If-None-Match condition per + Section 13.2 prior to performing the method.¶

+

+ To evaluate a received If-None-Match header field:¶

+
    +
  1. + If the field value is "*", the condition is false if the origin server + has a current representation for the target resource.¶ +
    2. +
  2. + If the field value is a list of entity tags, the condition is false if + one of the listed tags matches the entity tag of the selected representation.¶ +
    3. +
  3. + Otherwise, the condition is true.¶ +
    4. +
+

+ An origin server that evaluates an If-None-Match condition MUST NOT + perform the requested method if the condition evaluates to false; instead, + the origin server MUST respond with either + a) the 304 (Not Modified) status code if the request method + is GET or HEAD or b) the 412 (Precondition Failed) status + code for all other request methods.¶

+

+ Requirements on cache handling of a received If-None-Match header field + are defined in Section 4.3.2 of [CACHING].¶

+

+ Note that an If-None-Match header field with a list value containing "*" and + other values (including other instances of "*") is syntactically + invalid (therefore not allowed to be generated) and furthermore is + unlikely to be interoperable.¶

+
+
+
+
+

+13.1.3. If-Modified-Since +

+ + + +

+ The "If-Modified-Since" header field makes a GET or HEAD request method + conditional on the selected representation's modification + date being more + recent than the date provided in the field value. Transfer of the selected + representation's data is avoided if that data has not changed.¶

+ +
+
  If-Modified-Since = HTTP-date
+
¶ +
+

+ An example of the field is:¶

+
+
If-Modified-Since: Sat, 29 Oct 1994 19:43:31 GMT
+
¶ +
+

+ A recipient MUST ignore If-Modified-Since if the request contains an + If-None-Match header field; the condition in + If-None-Match is considered to be a more accurate + replacement for the condition in If-Modified-Since, and the two are only + combined for the sake of interoperating with older intermediaries that + might not implement If-None-Match.¶

+

+ A recipient MUST ignore the If-Modified-Since header field if the + received field value is not a valid HTTP-date, the field value has more than + one member, or if the request method is neither GET nor HEAD.¶

+

+ A recipient MUST ignore the If-Modified-Since header field if the + resource does not have a modification date available.¶

+

+ A recipient MUST interpret an If-Modified-Since field value's timestamp + in terms of the origin server's clock.¶

+

+ If-Modified-Since is typically used for two distinct purposes: + 1) to allow efficient updates of a cached representation that does not + have an entity tag and 2) to limit the scope of a web traversal to resources + that have recently changed.¶

+

+ When used for cache updates, a cache will typically use the value of the + cached message's Last-Modified header field to generate the field + value of If-Modified-Since. This behavior is most interoperable for cases + where clocks are poorly synchronized or when the server has chosen to only + honor exact timestamp matches (due to a problem with Last-Modified dates + that appear to go "back in time" when the origin server's clock is + corrected or a representation is restored from an archived backup). + However, caches occasionally generate the field value based on other data, + such as the Date header field of the cached message or the + clock time at which the message was received, particularly when the + cached message does not contain a Last-Modified header field.¶

+

+ When used for limiting the scope of retrieval to a recent time window, a + user agent will generate an If-Modified-Since field value based on either + its own clock or a Date header field received from the + server in a prior response. Origin servers that choose an exact + timestamp match based on the selected representation's + Last-Modified + header field will not be able to help the user agent limit its data + transfers to only those changed during the specified window.¶

+

+ When an origin server receives a request that selects a representation + and that request includes an If-Modified-Since header field without an + If-None-Match header field, the origin server SHOULD + evaluate the If-Modified-Since condition per + Section 13.2 prior to performing the method.¶

+

+ To evaluate a received If-Modified-Since header field:¶

+
    +
  1. + If the selected representation's last modification date is earlier or + equal to the date provided in the field value, the condition is false.¶ +
    2. +
  2. + Otherwise, the condition is true.¶ +
    3. +
+

+ An origin server that evaluates an If-Modified-Since condition + SHOULD NOT perform the requested method if the condition evaluates to + false; instead, + the origin server SHOULD generate a 304 (Not Modified) + response, including only those metadata that are useful for identifying or + updating a previously cached response.¶

+

+ Requirements on cache handling of a received If-Modified-Since header field + are defined in Section 4.3.2 of [CACHING].¶

+
+
+
+
+

+13.1.4. If-Unmodified-Since +

+ + + +

+ The "If-Unmodified-Since" header field makes the request method conditional + on the selected representation's last modification date being + earlier than or equal to the date provided in the field value. + This field accomplishes the + same purpose as If-Match for cases where the user agent does + not have an entity tag for the representation.¶

+ +
+
  If-Unmodified-Since = HTTP-date
+
¶ +
+

+ An example of the field is:¶

+
+
If-Unmodified-Since: Sat, 29 Oct 1994 19:43:31 GMT
+
¶ +
+

+ A recipient MUST ignore If-Unmodified-Since if the request contains an + If-Match header field; the condition in + If-Match is considered to be a more accurate replacement for + the condition in If-Unmodified-Since, and the two are only combined for the + sake of interoperating with older intermediaries that might not implement + If-Match.¶

+

+ A recipient MUST ignore the If-Unmodified-Since header field if the + received field value is not a valid HTTP-date (including when the field + value appears to be a list of dates).¶

+

+ A recipient MUST ignore the If-Unmodified-Since header field if the + resource does not have a modification date available.¶

+

+ A recipient MUST interpret an If-Unmodified-Since field value's timestamp + in terms of the origin server's clock.¶

+

+ If-Unmodified-Since is most often used with state-changing methods + (e.g., POST, PUT, DELETE) to prevent accidental overwrites when multiple + user agents might be acting in parallel on a resource that does + not supply entity tags with its representations (i.e., to prevent the + "lost update" problem). + In general, it can be used with any method that involves the selection + or modification of a representation to abort the request if the + selected representation's last modification date has + changed since the date provided in the If-Unmodified-Since field value.¶

+

+ When an origin server receives a request that selects a representation + and that request includes an If-Unmodified-Since header field without + an If-Match header field, + the origin server MUST evaluate the If-Unmodified-Since condition per + Section 13.2 prior to performing the method.¶

+

+ To evaluate a received If-Unmodified-Since header field:¶

+
    +
  1. + If the selected representation's last modification date is earlier than or + equal to the date provided in the field value, the condition is true.¶ +
    2. +
  2. + Otherwise, the condition is false.¶ +
    3. +
+

+ An origin server that evaluates an If-Unmodified-Since condition MUST NOT + perform the requested method if the condition evaluates to false. + Instead, the origin server MAY indicate that the conditional request + failed by responding with a 412 (Precondition Failed) + status code. Alternatively, if the request is a state-changing operation + that appears to have already been applied to the selected representation, + the origin server MAY respond with a 2xx (Successful) + status code + (i.e., the change requested by the user agent has already succeeded, but + the user agent might not be aware of it, perhaps because the prior response + was lost or an equivalent change was made by some other user agent).¶

+

+ Allowing an origin server to send a success response when a change request + appears to have already been applied is more efficient for many authoring + use cases, but comes with some risk if multiple user agents are making + change requests that are very similar but not cooperative. + In those cases, an origin server is better off being stringent in sending + 412 for every failed precondition on an unsafe method.¶

+

+ A client MAY send an If-Unmodified-Since header field in a + GET request to indicate that it would prefer a + 412 (Precondition Failed) response if the selected + representation has been modified. However, this is only useful in range + requests (Section 14) for completing a previously + received partial representation when there is no desire for a new + representation. If-Range (Section 13.1.5) + is better suited for range requests when the client prefers to receive a + new representation.¶

+

+ A cache or intermediary MAY ignore If-Unmodified-Since because its + interoperability features are only necessary for an origin server.¶

+
+
+
+
+

+13.1.5. If-Range +

+ + + +

+ The "If-Range" header field provides a special conditional request + mechanism that is similar to the If-Match and + If-Unmodified-Since header fields but that instructs the + recipient to ignore the Range header field if the validator + doesn't match, resulting in transfer of the new selected representation + instead of a 412 (Precondition Failed) response.¶

+

+ If a client has a partial copy of a representation and wishes + to have an up-to-date copy of the entire representation, it could use the + Range header field with a conditional GET (using + either or both of If-Unmodified-Since and + If-Match.) However, if the precondition fails because the + representation has been modified, the client would then have to make a + second request to obtain the entire current representation.¶

+

+ The "If-Range" header field allows a client to "short-circuit" the second + request. Informally, its meaning is as follows: if the representation is unchanged, + send me the part(s) that I am requesting in Range; otherwise, send me the + entire representation.¶

+ +
+
  If-Range = entity-tag / HTTP-date
+
¶ +
+

+ A valid entity-tag can be distinguished from a valid + HTTP-date by examining the first three characters for a + DQUOTE.¶

+

+ A client MUST NOT generate an If-Range header field in a request that + does not contain a Range header field. + A server MUST ignore an If-Range header field received in a request that + does not contain a Range header field. + An origin server MUST ignore an If-Range header field received in a + request for a target resource that does not support Range requests.¶

+

+ A client MUST NOT generate an If-Range header field containing an + entity tag that is marked as weak. + A client MUST NOT generate an If-Range header field containing an + HTTP-date unless the client has no entity tag for + the corresponding representation and the date is a strong validator + in the sense defined by Section 8.8.2.2.¶

+

+ A server that receives an If-Range header field on a Range request MUST + evaluate the condition per Section 13.2 prior to + performing the method.¶

+

+ To evaluate a received If-Range header field containing an + HTTP-date:¶

+
    +
  1. If the HTTP-date validator provided is not a + strong validator in the sense defined by + Section 8.8.2.2, the condition is false.¶ +
    2. +
  2. If the HTTP-date validator provided exactly matches + the Last-Modified field value for the selected + representation, the condition is true.¶ +
    3. +
  3. Otherwise, the condition is false.¶ +
    4. +
+

+ To evaluate a received If-Range header field containing an + entity-tag:¶

+
    +
  1. If the entity-tag validator provided exactly matches + the ETag field value for the selected representation + using the strong comparison function + (Section 8.8.3.2), the condition is true.¶ +
    2. +
  2. Otherwise, the condition is false.¶ +
    3. +
+

+ A recipient of an If-Range header field MUST ignore the + Range header field if the If-Range condition + evaluates to false. Otherwise, the recipient SHOULD process the + Range header field as requested.¶

+

+ Note that the If-Range comparison is by exact match, including when the + validator is an HTTP-date, and so it + differs from the "earlier than or equal to" comparison used when evaluating + an If-Unmodified-Since conditional.¶

+
+
+
+
+
+
+

+13.2. Evaluation of Preconditions +

+
+
+

+13.2.1. When to Evaluate +

+

+ Except when excluded below, a recipient cache or origin server MUST + evaluate received request preconditions after it has successfully performed + its normal request checks and just before it would process the request content + (if any) or perform the action associated with the request method. + A server MUST ignore all received preconditions if its response to the + same request without those conditions, prior to processing the request content, + would have been a status code other than a 2xx (Successful) + or 412 (Precondition Failed). + In other words, redirects and failures that can be detected before + significant processing occurs take precedence over the evaluation + of preconditions.¶

+

+ A server that is not the origin server for the target resource and cannot + act as a cache for requests on the target resource MUST NOT evaluate the + conditional request header fields defined by this specification, and it + MUST forward them if the request is forwarded, since the generating + client intends that they be evaluated by a server that can provide a + current representation. + Likewise, a server MUST ignore the conditional request header fields + defined by this specification when received with a request method that does + not involve the selection or modification of a + selected representation, such as CONNECT, OPTIONS, or TRACE.¶

+

+ Note that protocol extensions can modify the conditions under which + preconditions are evaluated or the consequences of their evaluation. + For example, the immutable cache directive + (defined by [RFC8246]) instructs caches to forgo + forwarding conditional requests when they hold a fresh response.¶

+

+ Although conditional request header fields are defined as being usable with + the HEAD method (to keep HEAD's semantics consistent with those of GET), + there is no point in sending a conditional HEAD because a successful + response is around the same size as a 304 (Not Modified) + response and more useful than a 412 (Precondition Failed) + response.¶

+
+
+
+
+

+13.2.2. Precedence of Preconditions +

+

+ When more than one conditional request header field is present in a request, + the order in which the fields are evaluated becomes important. In practice, + the fields defined in this document are consistently implemented in a + single, logical order, since "lost update" preconditions have more strict + requirements than cache validation, a validated cache is more efficient + than a partial response, and entity tags are presumed to be more accurate + than date validators.¶

+

+ A recipient cache or origin server MUST evaluate the request + preconditions defined by this specification in the following order:¶

+
    +
  1. +
    +

    When recipient is the origin server and + If-Match is present, + evaluate the If-Match precondition:¶

    + +
    +
    2. +
  2. +
    +

    When recipient is the origin server, + If-Match is not present, and + If-Unmodified-Since is present, + evaluate the If-Unmodified-Since precondition:¶

    + +
    +
    3. +
  3. +
    +

    When If-None-Match is present, + evaluate the If-None-Match precondition:¶

    + +
    +
    4. +
  4. +
    +

    When the method is GET or HEAD, + If-None-Match is not present, and + If-Modified-Since is present, + evaluate the If-Modified-Since precondition:¶

    + +
    +
    5. +
  5. +
    +

    When the method is GET and both + Range and If-Range are present, + evaluate the If-Range precondition:¶

    + +
    +
    6. +
  6. +
    +

    Otherwise,¶

    +
      +
    • perform the requested method and + respond according to its success or failure.¶ +
      • +
    +
    +
    7. +
+

+ Any extension to HTTP that defines additional conditional request + header fields ought to define the order + for evaluating such fields in relation to those defined in this document + and other conditionals that might be found in practice.¶

+
+
+
+
+
+
+
+
+

+14. Range Requests +

+

+ Clients often encounter interrupted data + transfers as a result of canceled requests or dropped connections. When a + client has stored a partial representation, it is desirable to request the + remainder of that representation in a subsequent request rather than + transfer the entire representation. Likewise, devices with limited local + storage might benefit from being able to request only a subset of a larger + representation, such as a single page of a very large document, or the + dimensions of an embedded image.¶

+

+ Range requests are an OPTIONAL feature + of HTTP, designed so that recipients not implementing this feature (or not + supporting it for the target resource) can respond as if it is a normal + GET request without impacting interoperability. Partial responses are + indicated by a distinct status code to not be mistaken for full responses + by caches that might not implement the feature.¶

+
+
+

+14.1. Range Units +

+

+ Representation data can be partitioned into subranges when there are + addressable structural units inherent to that data's content coding or + media type. For example, octet (a.k.a. byte) boundaries are a structural + unit common to all representation data, allowing partitions of the data to + be identified as a range of bytes at some offset from the start or end of + that data.¶

+

+ This general notion of a "range unit" is used + in the Accept-Ranges (Section 14.3) + response header field to advertise support for range requests, the + Range (Section 14.2) request header field + to delineate the parts of a representation that are requested, and the + Content-Range (Section 14.4) + header field to describe which part of a representation is being + transferred.¶

+ +
+
  range-unit       = token
+
¶ +
+

+ All range unit names are case-insensitive and ought to be registered + within the "HTTP Range Unit Registry", as defined in + Section 16.5.1.¶

+

+ Range units are intended to be extensible, as described in + Section 16.5.¶

+
+
+

+14.1.1. Range Specifiers +

+ + +

+ Ranges are expressed in terms of a range unit paired with a set of range + specifiers. The range unit name determines what kinds of range-spec + are applicable to its own specifiers. Hence, the following grammar is + generic: each range unit is expected to specify requirements on when + int-range, suffix-range, and + other-range are allowed.¶

+
+

+ + + + A range request can specify a single range or a set + of ranges within a single representation.¶

+
+ + + +
+
  ranges-specifier = range-unit "=" range-set
+  range-set        = 1#range-spec
+  range-spec       = int-range
+                   / suffix-range
+                   / other-range
+
¶ +
+
+

+ + + + An int-range is a range expressed as two non-negative + integers or as one non-negative integer through to the end of the + representation data. + The range unit specifies what the integers mean (e.g., they might indicate + unit offsets from the beginning, inclusive numbered parts, etc.).¶

+
+ + + +
+
  int-range     = first-pos "-" [ last-pos ]
+  first-pos     = 1*DIGIT
+  last-pos      = 1*DIGIT
+
¶ +
+

+ An int-range is invalid if the + last-pos value is present and less than the + first-pos.¶

+
+

+ + + A suffix-range is a range expressed as a suffix of the + representation data with the provided non-negative integer maximum length + (in range units). In other words, the last N units of the representation + data.¶

+
+ + +
+
  suffix-range  = "-" suffix-length
+  suffix-length = 1*DIGIT
+
¶ +
+
+

+ + To provide for extensibility, the other-range rule is a + mostly unconstrained grammar that allows application-specific or future + range units to define additional range specifiers.¶

+
+ +
+
  other-range   = 1*( %x21-2B / %x2D-7E )
+                ; 1*(VCHAR excluding comma)
+
¶ +
+

+ A ranges-specifier is invalid if it contains any + range-spec that is invalid or undefined for the indicated + range-unit.¶

+
+

+ A valid ranges-specifier is "satisfiable" + if it contains at least one range-spec that is + satisfiable, as defined by the indicated range-unit. + Otherwise, the ranges-specifier is + "unsatisfiable".¶

+
+
+
+
+
+

+14.1.2. Byte Ranges +

+

+ The "bytes" range unit is used to express subranges of a representation + data's octet sequence. + Each byte range is expressed as an integer range at some offset, relative + to either the beginning (int-range) or end + (suffix-range) of the representation data. + Byte ranges do not use the other-range specifier.¶

+

+ The first-pos value in a bytes int-range + gives the offset of the first byte in a range. + The last-pos value gives the offset of the last + byte in the range; that is, the byte positions specified are inclusive. + Byte offsets start at zero.¶

+

+ If the representation data has a content coding applied, each byte range is + calculated with respect to the encoded sequence of bytes, not the sequence + of underlying bytes that would be obtained after decoding.¶

+

+ Examples of bytes range specifiers:¶

+
    +
  • +

    The first 500 bytes (byte offsets 0-499, inclusive):¶

    +
    +
    +     bytes=0-499
+
    ¶ +
    +
    • +
  • +

    The second 500 bytes (byte offsets 500-999, inclusive):¶

    +
    +
    +     bytes=500-999
+
    ¶ +
    +
    • +
+

+ A client can limit the number of bytes requested without knowing the size + of the selected representation. + If the last-pos value is absent, or if the value is + greater than or equal to the current length of the representation data, the + byte range is interpreted as the remainder of the representation (i.e., the + server replaces the value of last-pos with a value that + is one less than the current length of the selected representation).¶

+

+ A client can refer to the last N bytes (N > 0) of the selected + representation using a suffix-range. + If the selected representation is shorter than the specified + suffix-length, the entire representation is used.¶

+

+ Additional examples, assuming a representation of length 10000:¶

+
    +
  • +

    The final 500 bytes (byte offsets 9500-9999, inclusive):¶

    +
    +
    +     bytes=-500
+
    ¶ +
    +

    Or:¶

    +
    +
    +     bytes=9500-
+
    ¶ +
    +
    • +
  • +

    The first and last bytes only (bytes 0 and 9999):¶

    +
    +
    +     bytes=0-0,-1
+
    ¶ +
    +
    • +
  • +

    The first, middle, and last 1000 bytes:¶

    +
    +
    +     bytes= 0-999, 4500-5499, -1000
+
    ¶ +
    +
    • +
  • +

    Other valid (but not canonical) specifications of the second 500 + bytes (byte offsets 500-999, inclusive):¶

    +
    +
    +     bytes=500-600,601-999
+     bytes=500-700,601-999
+
    ¶ +
    +
    • +
+

+ For a GET request, a valid bytes range-spec + is satisfiable if it is either:¶

+ +

+ When a selected representation has zero length, the only + satisfiable form of range-spec in a + GET request is a suffix-range with a + non-zero suffix-length.¶

+

+ In the byte-range syntax, first-pos, + last-pos, and suffix-length are + expressed as decimal number of octets. Since there is no predefined limit + to the length of content, recipients MUST anticipate potentially + large decimal numerals and prevent parsing errors due to integer conversion + overflows.¶

+
+
+
+
+
+
+

+14.2. Range +

+ + + +

+ The "Range" header field on a GET request modifies the method semantics to + request transfer of only one or more subranges of the + selected representation data (Section 8.1), + rather than the entire selected representation.¶

+ +
+
  Range = ranges-specifier
+
¶ +
+

+ A server MAY ignore the Range header field. However, origin servers and + intermediate caches ought to support byte ranges when possible, since they + support efficient recovery from partially failed transfers and partial + retrieval of large representations.¶

+

+ A server MUST ignore a Range header field received with a request method + that is unrecognized or for which range handling is not defined. For this + specification, GET is the only method for which range handling + is defined.¶

+

+ An origin server MUST ignore a Range header field that contains a range + unit it does not understand. A proxy MAY discard a Range header + field that contains a range unit it does not understand.¶

+

+ A server that supports range requests MAY ignore or reject a + Range header field that contains an invalid + ranges-specifier (Section 14.1.1), + a ranges-specifier with more than two overlapping ranges, + or a set of many small ranges that are not listed in ascending order, + since these are indications of either a broken client or a deliberate + denial-of-service attack (Section 17.15). + A client SHOULD NOT request multiple ranges that are inherently less + efficient to process and transfer than a single range that encompasses the + same data.¶

+

+ A server that supports range requests MAY ignore a Range + header field when the selected representation has no content + (i.e., the selected representation's data is of zero length).¶

+

+ A client that is requesting multiple ranges SHOULD list those ranges in + ascending order (the order in which they would typically be received in a + complete representation) unless there is a specific need to request a later + part earlier. For example, a user agent processing a large representation + with an internal catalog of parts might need to request later parts first, + particularly if the representation consists of pages stored in reverse + order and the user agent wishes to transfer one page at a time.¶

+

+ The Range header field is evaluated after evaluating the precondition header + fields defined in Section 13.1, and only if the result in absence + of the Range header field would be a 200 (OK) response. In + other words, Range is ignored when a conditional GET would result in a + 304 (Not Modified) response.¶

+

+ The If-Range header field (Section 13.1.5) can be used as + a precondition to applying the Range header field.¶

+

+ If all of the preconditions are true, the server supports the Range header + field for the target resource, the received Range field-value contains a + valid ranges-specifier with a range-unit + supported for that target resource, and that + ranges-specifier is satisfiable with respect + to the selected representation, + the server SHOULD send a 206 (Partial Content) response + with content containing one or more partial representations + that correspond to the satisfiable range-spec(s) requested.¶

+

+ The above does not imply that a server will send all requested ranges. + In some cases, it may only be possible (or efficient) to send a portion of + the requested ranges first, while expecting the client to re-request the + remaining portions later if they are still desired + (see Section 15.3.7).¶

+

+ If all of the preconditions are true, the server supports the Range header + field for the target resource, the received Range field-value contains a + valid ranges-specifier, and either the + range-unit is not supported for that target resource or + the ranges-specifier is unsatisfiable with respect to + the selected representation, the server SHOULD send a + 416 (Range Not Satisfiable) response.¶

+
+
+
+
+

+14.3. Accept-Ranges +

+ + + +

+ The "Accept-Ranges" field in a response indicates whether an upstream + server supports range requests for the target resource.¶

+ + +
+
  Accept-Ranges     = acceptable-ranges
+  acceptable-ranges = 1#range-unit
+
¶ +
+

+ For example, a server that supports + byte-range requests (Section 14.1.2) can send the field¶

+
+
Accept-Ranges: bytes
+
¶ +
+

+ to indicate that it supports byte range requests for that target resource, + thereby encouraging its use by the client for future partial requests on + the same request path. + Range units are defined in Section 14.1.¶

+

+ A client MAY generate range requests regardless of having received an + Accept-Ranges field. The information only provides advice for the sake of + improving performance and reducing unnecessary network transfers.¶

+

+ Conversely, a client MUST NOT assume that receiving an Accept-Ranges field + means that future range requests will return partial responses. The content might + change, the server might only support range requests at certain times or under + certain conditions, or a different intermediary might process the next request.¶

+

+ A server that does not support any kind of range request for the target + resource MAY send¶

+
+
Accept-Ranges: none
+
¶ +
+

+ to advise the client not to attempt a range request on the same request path. + The range unit "none" is reserved for this purpose.¶

+

+ The Accept-Ranges field MAY be sent in a trailer section, but is preferred + to be sent as a header field because the information is particularly useful + for restarting large information transfers that have failed in mid-content + (before the trailer section is received).¶

+
+
+
+
+

+14.4. Content-Range +

+ + + +

+ The "Content-Range" header field is sent in a single part + 206 (Partial Content) response to indicate the partial range + of the selected representation enclosed as the message content, sent in + each part of a multipart 206 response to indicate the range enclosed within + each body part (Section 14.6), and sent in 416 (Range Not Satisfiable) + responses to provide information about the selected representation.¶

+ + + + + + + +
+
  Content-Range       = range-unit SP
+                        ( range-resp / unsatisfied-range )
+
+  range-resp          = incl-range "/" ( complete-length / "*" )
+  incl-range          = first-pos "-" last-pos
+  unsatisfied-range   = "*/" complete-length
+
+  complete-length     = 1*DIGIT
+
¶ +
+

+ If a 206 (Partial Content) response contains a + Content-Range header field with a range unit + (Section 14.1) that the recipient does not understand, the + recipient MUST NOT attempt to recombine it with a stored representation. + A proxy that receives such a message SHOULD forward it downstream.¶

+

+ Content-Range might also be sent as a request modifier to request a + partial PUT, as described in Section 14.5, based on private + agreements between client and origin server. + A server MUST ignore a Content-Range header field received in a request + with a method for which Content-Range support is not defined.¶

+

+ For byte ranges, a sender SHOULD indicate the complete length of the + representation from which the range has been extracted, unless the complete + length is unknown or difficult to determine. An asterisk character ("*") in + place of the complete-length indicates that the representation length was + unknown when the header field was generated.¶

+

+ The following example illustrates when the complete length of the selected + representation is known by the sender to be 1234 bytes:¶

+
+
Content-Range: bytes 42-1233/1234
+
¶ +
+

+ and this second example illustrates when the complete length is unknown:¶

+
+
Content-Range: bytes 42-1233/*
+
¶ +
+

+ A Content-Range field value is invalid if it contains a + range-resp that has a last-pos + value less than its first-pos value, or a + complete-length value less than or equal to its + last-pos value. The recipient of an invalid + Content-Range + MUST NOT attempt to recombine the received + content with a stored representation.¶

+

+ A server generating a 416 (Range Not Satisfiable) response + to a byte-range request SHOULD send a Content-Range header field with an + unsatisfied-range value, as in the following example:¶

+
+
Content-Range: bytes */1234
+
¶ +
+

+ The complete-length in a 416 response indicates the current length of the + selected representation.¶

+

+ The Content-Range header field has no meaning for status codes that do + not explicitly describe its semantic. For this specification, only the + 206 (Partial Content) and + 416 (Range Not Satisfiable) status codes describe a meaning + for Content-Range.¶

+

+ The following are examples of Content-Range values in which the + selected representation contains a total of 1234 bytes:¶

+
    +
  • +

    The first 500 bytes:¶

    +
    +
    Content-Range: bytes 0-499/1234
+
    ¶ +
    +
    • +
  • +

    The second 500 bytes:¶

    +
    +
    Content-Range: bytes 500-999/1234
+
    ¶ +
    +
    • +
  • +

    All except for the first 500 bytes:¶

    +
    +
    Content-Range: bytes 500-1233/1234
+
    ¶ +
    +
    • +
  • +

    The last 500 bytes:¶

    +
    +
    Content-Range: bytes 734-1233/1234
+
    ¶ +
    +
    • +
+
+
+
+
+

+14.5. Partial PUT +

+ + + +

+ Some origin servers support PUT of a partial representation + when the user agent sends a Content-Range header field + (Section 14.4) in the request, though + such support is inconsistent and depends on private agreements with + user agents. In general, it requests that the state of the + target resource be partly replaced with the enclosed content + at an offset and length indicated by the Content-Range value, where the + offset is relative to the current selected representation.¶

+

+ An origin server SHOULD respond with a 400 (Bad Request) + status code if it receives Content-Range on a PUT for a + target resource that does not support partial PUT requests.¶

+

+ Partial PUT is not backwards compatible with the original definition of PUT. + It may result in the content being written as a complete replacement for the + current representation.¶

+

+ Partial resource updates are also possible by targeting a separately + identified resource with state that overlaps or extends a portion of the + larger resource, or by using a different method that has been specifically + defined for partial updates (for example, the PATCH method defined in + [RFC5789]).¶

+
+
+
+
+

+14.6. Media Type multipart/byteranges +

+ + +

+ When a 206 (Partial Content) response message includes the + content of multiple ranges, they are transmitted as body parts in a + multipart message body ([RFC2046], Section 5.1) + with the media type of "multipart/byteranges".¶

+

+ The "multipart/byteranges" media type includes one or more body parts, each + with its own Content-Type and Content-Range + fields. The required boundary parameter specifies the boundary string used + to separate each body part.¶

+

+ Implementation Notes:¶

+
    +
  1. Additional CRLFs might precede the first boundary string in the body.¶ +
    2. +
  2. Although [RFC2046] permits the boundary string to be + quoted, some existing implementations handle a quoted boundary + string incorrectly.¶ +
    3. +
  3. A number of clients and servers were coded to an early draft + of the byteranges specification that used a media type of + "multipart/x-byteranges", + which is almost (but not quite) compatible with this type.¶ +
    4. +
+

+ Despite the name, the "multipart/byteranges" media type is not limited to + byte ranges. The following example uses an "exampleunit" range unit:¶

+
+
HTTP/1.1 206 Partial Content
+Date: Tue, 14 Nov 1995 06:25:24 GMT
+Last-Modified: Tue, 14 July 04:58:08 GMT
+Content-Length: 2331785
+Content-Type: multipart/byteranges; boundary=THIS_STRING_SEPARATES
+
+--THIS_STRING_SEPARATES
+Content-Type: video/example
+Content-Range: exampleunit 1.2-4.3/25
+
+...the first range...
+--THIS_STRING_SEPARATES
+Content-Type: video/example
+Content-Range: exampleunit 11.2-14.3/25
+
+...the second range
+--THIS_STRING_SEPARATES--
+
¶ +
+

+ The following information serves as the registration form for the + "multipart/byteranges" media type.¶

+
+
Type name:
+
multipart¶ +
+
+
Subtype name:
+
byteranges¶ +
+
+
Required parameters:
+
boundary¶ +
+
+
Optional parameters:
+
N/A¶ +
+
+
Encoding considerations:
+
only "7bit", "8bit", or "binary" are permitted¶ +
+
+
Security considerations:
+
see Section 17¶ +
+
+
Interoperability considerations:
+
N/A¶ +
+
+
Published specification:
+
RFC 9110 (see Section 14.6)¶ +
+
+
Applications that use this media type:
+
HTTP components supporting multiple ranges in a single request¶ +
+
+
Fragment identifier considerations:
+
N/A¶ +
+
+
Additional information:
+
+
+
Deprecated alias names for this type:
+
N/A¶ +
+
+
Magic number(s):
+
N/A¶ +
+
+
File extension(s):
+
N/A¶ +
+
+
Macintosh file type code(s):
+
N/A¶ +
+
+
+
+
+
Person and email address to contact for further information:
+
See Authors' Addresses section.¶ +
+
+
Intended usage:
+
COMMON¶ +
+
+
Restrictions on usage:
+
N/A¶ +
+
+
Author:
+
See Authors' Addresses section.¶ +
+
+
Change controller:
+
IESG¶ +
+
+
+
+
+
+
+
+
+

+15. Status Codes +

+ +

+ The status code of a response is a three-digit integer code that describes + the result of the request and the semantics of the response, including + whether the request was successful and what content is enclosed (if any). + All valid status codes are within the range of 100 to 599, inclusive.¶

+

+ The first digit of the status code defines the class of response. The + last two digits do not have any categorization role. There are five + values for the first digit:¶

+ +

+ HTTP status codes are extensible. A client is not required to understand + the meaning of all registered status codes, though such understanding is + obviously desirable. However, a client MUST understand the class of any + status code, as indicated by the first digit, and treat an unrecognized + status code as being equivalent to the x00 status code of that class.¶

+

+ For example, if a client receives an unrecognized status code of 471, + it can see from the first digit that there was something wrong with its + request and treat the response as if it had received a + 400 (Bad Request) status code. The response + message will usually contain a representation that explains the status.¶

+

+ Values outside the range 100..599 are invalid. Implementations often use + three-digit integer values outside of that range (i.e., 600..999) for + internal communication of non-HTTP status (e.g., library errors). A client + that receives a response with an invalid status code SHOULD process the + response as if it had a 5xx (Server Error) status code.¶

+
+

+ + + + A single request can have multiple associated responses: zero or more + "interim" (non-final) responses with status codes in the + "informational" (1xx) range, followed by exactly one + "final" response with a status code in one of the other ranges.¶

+
+
+
+

+15.1. Overview of Status Codes +

+

+ The status codes listed below are defined in this specification. + The reason phrases listed here are only recommendations -- they can be + replaced by local equivalents or left out altogether without affecting the + protocol.¶

+

+ Responses with status codes that are defined as heuristically cacheable + (e.g., 200, 203, 204, 206, 300, 301, 308, 404, 405, 410, 414, and 501 in this + specification) can be reused by a cache with heuristic expiration unless + otherwise indicated by the method definition or explicit cache controls + [CACHING]; all other status codes are not heuristically cacheable.¶

+

+ Additional status codes, outside the scope of this specification, have been + specified for use in HTTP. All such status codes ought to be registered + within the "Hypertext Transfer Protocol (HTTP) Status Code Registry", + as described in Section 16.2.¶

+
+
+
+
+

+15.2. Informational 1xx +

+ + +

+ The 1xx (Informational) class of status code indicates an + interim response for communicating connection status or request progress + prior to completing the requested action and sending a final response. + Since HTTP/1.0 did not define any 1xx status codes, a server MUST NOT send + a 1xx response to an HTTP/1.0 client.¶

+

+ A 1xx response is terminated by the end of the header section; + it cannot contain content or trailers.¶

+

+ A client MUST be able to parse one or more 1xx responses received + prior to a final response, even if the client does not expect one. + A user agent MAY ignore unexpected 1xx responses.¶

+

+ A proxy MUST forward 1xx responses unless the proxy itself + requested the generation of the 1xx response. For example, if a + proxy adds an "Expect: 100-continue" header field when it forwards a request, + then it need not forward the corresponding 100 (Continue) + response(s).¶

+
+
+

+15.2.1. 100 Continue +

+ +

+ The 100 (Continue) status code indicates that the initial + part of a request has been received and has not yet been rejected by the + server. The server intends to send a final response after the request has + been fully received and acted upon.¶

+

+ When the request contains an Expect header field that + includes a 100-continue expectation, the 100 response + indicates that the server wishes to receive the request content, + as described in Section 10.1.1. The client + ought to continue sending the request and discard the 100 response.¶

+

+ If the request did not contain an Expect header field + containing the 100-continue expectation, + the client can simply discard this interim response.¶

+
+
+
+
+

+15.2.2. 101 Switching Protocols +

+ +

+ The 101 (Switching Protocols) status code indicates that the + server understands and is willing to comply with the client's request, + via the Upgrade header field (Section 7.8), for + a change in the application protocol being used on this connection. + The server MUST generate an Upgrade header field in the response that + indicates which protocol(s) will be in effect after this response.¶

+

+ It is assumed that the server will only agree to switch protocols when + it is advantageous to do so. For example, switching to a newer version of + HTTP might be advantageous over older versions, and switching to a + real-time, synchronous protocol might be advantageous when delivering + resources that use such features.¶

+
+
+
+
+
+
+

+15.3. Successful 2xx +

+ + +

+ The 2xx (Successful) class of status code indicates that + the client's request was successfully received, understood, and accepted.¶

+
+
+

+15.3.1. 200 OK +

+ +

+ The 200 (OK) status code indicates that the request has + succeeded. The content sent in a 200 response depends on the request + method. For the methods defined by this specification, the intended meaning + of the content can be summarized as:¶

+ + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +
Table 6
Request MethodResponse content is a representation of:
GETthe target resource +
HEADthe target resource, like GET, but without + transferring the representation data
POSTthe status of, or results obtained from, the action
PUT, DELETEthe status of the action
OPTIONScommunication options for the target resource
TRACEthe request message as received by the server returning the + trace
+

+ Aside from responses to CONNECT, a 200 response is expected to contain + message content unless the message framing explicitly indicates that the + content has zero length. If some aspect of the request indicates a + preference for no content upon success, the origin server ought to send a + 204 (No Content) response instead. + For CONNECT, there is no content because the successful result is a + tunnel, which begins immediately after the 200 response header section.¶

+

+ A 200 response is heuristically cacheable; i.e., unless otherwise indicated by + the method definition or explicit cache controls (see Section 4.2.2 of [CACHING]).¶

+

+ In 200 responses to GET or HEAD, an origin server SHOULD send any + available validator fields (Section 8.8) for the + selected representation, with both a strong entity tag and + a Last-Modified date being preferred.¶

+

+ In 200 responses to state-changing methods, any validator fields + (Section 8.8) sent in the response convey the + current validators for the new representation formed as a result of + successfully applying the request semantics. Note that the PUT method + (Section 9.3.4) has additional requirements that might preclude + sending such validators.¶

+
+
+
+
+

+15.3.2. 201 Created +

+ +

+ The 201 (Created) status code indicates that the request has + been fulfilled and has resulted in one or more new resources being created. + The primary resource created by the request is identified by either a + Location header field in the response or, if no + Location header field is received, by the target URI.¶

+

+ The 201 response content typically describes and links to the resource(s) + created. Any validator fields (Section 8.8) + sent in the response convey the current validators for a new + representation created by the request. Note that the PUT method + (Section 9.3.4) has additional requirements that might preclude + sending such validators.¶

+
+
+
+
+

+15.3.3. 202 Accepted +

+ +

+ The 202 (Accepted) status code indicates that the request + has been accepted for processing, but the processing has not been + completed. The request might or might not eventually be acted upon, as it + might be disallowed when processing actually takes place. There is no + facility in HTTP for re-sending a status code from an asynchronous + operation.¶

+

+ The 202 response is intentionally noncommittal. Its purpose is to + allow a server to accept a request for some other process (perhaps a + batch-oriented process that is only run once per day) without + requiring that the user agent's connection to the server persist + until the process is completed. The representation sent with this + response ought to describe the request's current status and point to + (or embed) a status monitor that can provide the user with an estimate of + when the request will be fulfilled.¶

+
+
+
+
+

+15.3.4. 203 Non-Authoritative Information +

+ +

+ The 203 (Non-Authoritative Information) status code + indicates that the request was successful but the enclosed content has been + modified from that of the origin server's 200 (OK) response + by a transforming proxy (Section 7.7). This status code allows the + proxy to notify recipients when a transformation has been applied, since + that knowledge might impact later decisions regarding the content. For + example, future cache validation requests for the content might only be + applicable along the same request path (through the same proxies).¶

+

+ A 203 response is heuristically cacheable; i.e., unless otherwise indicated by + the method definition or explicit cache controls (see Section 4.2.2 of [CACHING]).¶

+
+
+
+
+

+15.3.5. 204 No Content +

+ +

+ The 204 (No Content) status code indicates that the server + has successfully fulfilled the request and that there is no additional + content to send in the response content. Metadata in the response + header fields refer to the target resource and its + selected representation after the requested action was applied.¶

+

+ For example, if a 204 status code is received in response to a PUT + request and the response contains an ETag field, then + the PUT was successful and the ETag field value contains the entity tag for + the new representation of that target resource.¶

+

+ The 204 response allows a server to indicate that the action has been + successfully applied to the target resource, while implying that the + user agent does not need to traverse away from its current "document view" + (if any). The server assumes that the user agent will provide some + indication of the success to its user, in accord with its own interface, + and apply any new or updated metadata in the response to its active + representation.¶

+

+ For example, a 204 status code is commonly used with document editing + interfaces corresponding to a "save" action, such that the document + being saved remains available to the user for editing. It is also + frequently used with interfaces that expect automated data transfers + to be prevalent, such as within distributed version control systems.¶

+

+ A 204 response is terminated by the end of the header section; + it cannot contain content or trailers.¶

+

+ A 204 response is heuristically cacheable; i.e., unless otherwise indicated by + the method definition or explicit cache controls (see Section 4.2.2 of [CACHING]).¶

+
+
+
+
+

+15.3.6. 205 Reset Content +

+ +

+ The 205 (Reset Content) status code indicates that the + server has fulfilled the request and desires that the user agent reset the + "document view", which caused the request to be sent, to its original state + as received from the origin server.¶

+

+ This response is intended to support a common data entry use case where + the user receives content that supports data entry (a form, notepad, + canvas, etc.), enters or manipulates data in that space, causes the entered + data to be submitted in a request, and then the data entry mechanism is + reset for the next entry so that the user can easily initiate another + input action.¶

+

+ Since the 205 status code implies that no additional content will be + provided, a server MUST NOT generate content in a 205 response.¶

+
+
+
+
+

+15.3.7. 206 Partial Content +

+ +

+ The 206 (Partial Content) status code indicates that the + server is successfully fulfilling a range request for the target resource + by transferring one or more parts of the + selected representation.¶

+

+ A server that supports range requests (Section 14) will + usually attempt to satisfy all of the requested ranges, since sending + less data will likely result in another client request for the remainder. + However, a server might want to send only a subset of the data requested + for reasons of its own, such as temporary unavailability, cache efficiency, + load balancing, etc. Since a 206 response is self-descriptive, the client + can still understand a response that only partially satisfies its range + request.¶

+

+ A client MUST inspect a 206 response's Content-Type and + Content-Range field(s) to determine what parts are enclosed + and whether additional requests are needed.¶

+

+ A server that generates a 206 response MUST generate the following + header fields, in addition to those required in the subsections below, + if the field would + have been sent in a 200 (OK) response to the same request: + Date, Cache-Control, ETag, + Expires, Content-Location, and + Vary.¶

+

+ A Content-Length header field present in a 206 response + indicates the number of octets in the content of this message, which is + usually not the complete length of the selected representation. + Each Content-Range header field includes information about the + selected representation's complete length.¶

+

+ A sender that generates a 206 response to a request with an If-Range + header field SHOULD NOT generate other representation header + fields beyond those required because the client + already has a prior response containing those header fields. + Otherwise, a sender MUST generate all of the representation header + fields that would have been sent in a 200 (OK) response + to the same request.¶

+

+ A 206 response is heuristically cacheable; i.e., unless otherwise indicated by + explicit cache controls (see Section 4.2.2 of [CACHING]).¶

+
+
+
+15.3.7.1. Single Part +
+

+ If a single part is being transferred, the server generating the 206 + response MUST generate a Content-Range header field, + describing what range of the selected representation is enclosed, and a + content consisting of the range. For example:¶

+
+
HTTP/1.1 206 Partial Content
+Date: Wed, 15 Nov 1995 06:25:24 GMT
+Last-Modified: Wed, 15 Nov 1995 04:58:08 GMT
+Content-Range: bytes 21010-47021/47022
+Content-Length: 26012
+Content-Type: image/gif
+
+... 26012 bytes of partial image data ...
+
¶ +
+
+
+
+
+
+15.3.7.2. Multiple Parts +
+

+ If multiple parts are being transferred, the server generating the 206 + response MUST generate "multipart/byteranges" content, as defined + in Section 14.6, and a + Content-Type header field containing the + "multipart/byteranges" media type and its required boundary parameter. + To avoid confusion with single-part responses, a server MUST NOT generate + a Content-Range header field in the HTTP header section of a + multiple part response (this field will be sent in each part instead).¶

+

+ Within the header area of each body part in the multipart content, the + server MUST generate a Content-Range header field + corresponding to the range being enclosed in that body part. + If the selected representation would have had a Content-Type + header field in a 200 (OK) response, the server SHOULD + generate that same Content-Type header field in the header area of + each body part. For example:¶

+
+
HTTP/1.1 206 Partial Content
+Date: Wed, 15 Nov 1995 06:25:24 GMT
+Last-Modified: Wed, 15 Nov 1995 04:58:08 GMT
+Content-Length: 1741
+Content-Type: multipart/byteranges; boundary=THIS_STRING_SEPARATES
+
+--THIS_STRING_SEPARATES
+Content-Type: application/pdf
+Content-Range: bytes 500-999/8000
+
+...the first range...
+--THIS_STRING_SEPARATES
+Content-Type: application/pdf
+Content-Range: bytes 7000-7999/8000
+
+...the second range
+--THIS_STRING_SEPARATES--
+
¶ +
+

+ When multiple ranges are requested, a server MAY coalesce any of the + ranges that overlap, or that are separated by a gap that is smaller than the + overhead of sending multiple parts, regardless of the order in which the + corresponding range-spec appeared in the received Range + header field. Since the typical overhead between each part of a + "multipart/byteranges" is around 80 bytes, depending on the selected + representation's media type and the chosen boundary parameter length, it + can be less efficient to transfer many small disjoint parts than it is to + transfer the entire selected representation.¶

+

+ A server MUST NOT generate a multipart response to a request for a single + range, since a client that does not request multiple parts might not + support multipart responses. However, a server MAY generate a + "multipart/byteranges" response with only a single body part if multiple + ranges were requested and only one range was found to be satisfiable or + only one range remained after coalescing. + A client that cannot process a "multipart/byteranges" response MUST NOT + generate a request that asks for multiple ranges.¶

+

+ A server that generates a multipart response SHOULD send + the parts in the same order that the corresponding range-spec appeared + in the received Range header field, excluding those ranges + that were deemed unsatisfiable or that were coalesced into other ranges. + A client that receives a multipart response MUST inspect the + Content-Range header field present in each body part in + order to determine which range is contained in that body part; a client + cannot rely on receiving the same ranges that it requested, nor the same + order that it requested.¶

+
+
+
+
+
+15.3.7.3. Combining Parts +
+

+ A response might transfer only a subrange of a representation if the + connection closed prematurely or if the request used one or more Range + specifications. After several such transfers, a client might have + received several ranges of the same representation. These ranges can only + be safely combined if they all have in common the same strong validator + (Section 8.8.1).¶

+

+ A client that has received multiple partial responses to GET requests on a + target resource MAY combine those responses into a larger continuous + range if they share the same strong validator.¶

+

+ If the most recent response is an incomplete 200 (OK) + response, then the header fields of that response are used for any + combined response and replace those of the matching stored responses.¶

+

+ If the most recent response is a 206 (Partial Content) + response and at least one of the matching stored responses is a + 200 (OK), then the combined response header fields consist + of the most recent 200 response's header fields. If all of the matching + stored responses are 206 responses, then the stored response with the most + recent header fields is used as the source of header fields for the + combined response, except that the client MUST use other header fields + provided in the new response, aside from Content-Range, to + replace all instances of the corresponding header fields in the stored + response.¶

+

+ The combined response content consists of the union of partial content + ranges within the new response and all of the matching stored responses. + If the union consists of the entire range of the representation, then the + client MUST process the combined response as if it were a complete + 200 (OK) response, including a Content-Length + header field that reflects the complete length. + Otherwise, the client MUST process the set of continuous ranges as one of + the following: + an incomplete 200 (OK) response if the combined response is + a prefix of the representation, + a single 206 (Partial Content) response containing + "multipart/byteranges" content, or + multiple 206 (Partial Content) responses, each with one + continuous range that is indicated by a Content-Range header + field.¶

+
+
+
+
+
+
+
+
+

+15.4. Redirection 3xx +

+ + +

+ The 3xx (Redirection) class of status code indicates that + further action needs to be taken by the user agent in order to fulfill the + request. There are several types of redirects:¶

+
    +
  1. + Redirects that indicate this resource might be available at a + different URI, as provided by the Location header field, + as in the status codes 301 (Moved Permanently), + 302 (Found), 307 (Temporary Redirect), and + 308 (Permanent Redirect).¶ +
    2. +
  2. + Redirection that offers a choice among matching resources capable + of representing this resource, as in the + 300 (Multiple Choices) status code.¶ +
    3. +
  3. + Redirection to a different resource, identified by the + Location header field, that can represent an indirect + response to the request, as in the 303 (See Other) + status code.¶ +
    4. +
  4. + Redirection to a previously stored result, as in the + 304 (Not Modified) status code.¶ +
    5. +
+ +

+ If a Location header field + (Section 10.2.2) is provided, the user agent MAY + automatically redirect its request to the URI referenced by the Location + field value, even if the specific status code is not understood. + Automatic redirection needs to be done with care for methods not known to be + safe, as defined in Section 9.2.1, since + the user might not wish to redirect an unsafe request.¶

+

+ When automatically following a redirected request, the user agent SHOULD + resend the original request message with the following modifications:¶

+
    +
  1. +

    + Replace the target URI with the URI referenced by the redirection response's + Location header field value after resolving it relative to the original + request's target URI.¶

    +
    2. +
  2. +

    + Remove header fields that were automatically generated by the implementation, + replacing them with updated values as appropriate to the new request. This + includes:¶

    +
      +
    1. Connection-specific header fields (see Section 7.6.1),¶ +
      2. +
    2. Header fields specific to the client's proxy configuration, + including (but not limited to) Proxy-Authorization,¶ +
      3. +
    3. Origin-specific header fields (if any), including (but not + limited to) Host,¶ +
      4. +
    4. Validating header fields that were added by the implementation's + cache (e.g., If-None-Match, + If-Modified-Since), and¶ +
      5. +
    5. Resource-specific header fields, including (but not limited to) + Referer, Origin, + Authorization, and Cookie.¶ +
      6. +
    +
    3. +
  3. +

    + Consider removing header fields that were not automatically generated by the + implementation (i.e., those present in the request because they were added + by the calling context) where there are security implications; this + includes but is not limited to Authorization and Cookie.¶

    +
    4. +
  4. +

    + Change the request method according to the redirecting status code's + semantics, if applicable.¶

    +
    5. +
  5. +

    + If the request method has been changed to GET or HEAD, remove + content-specific header fields, including (but not limited to) + Content-Encoding, + Content-Language, Content-Location, + Content-Type, Content-Length, + Digest, Last-Modified.¶

    +
    6. +
+

+ A client SHOULD detect and intervene in cyclical redirections (i.e., + "infinite" redirection loops).¶

+ +
+
+

+15.4.1. 300 Multiple Choices +

+ +

+ The 300 (Multiple Choices) status code indicates that the + target resource has more than one representation, each with + its own more specific identifier, and information about the alternatives is + being provided so that the user (or user agent) can select a preferred + representation by redirecting its request to one or more of those + identifiers. In other words, the server desires that the user agent engage + in reactive negotiation to select the most appropriate representation(s) + for its needs (Section 12).¶

+

+ If the server has a preferred choice, the server SHOULD generate a + Location header field containing a preferred choice's URI + reference. The user agent MAY use the Location field value for automatic + redirection.¶

+

+ For request methods other than HEAD, the server SHOULD generate content + in the 300 response containing a list of representation metadata and URI + reference(s) from which the user or user agent can choose the one most + preferred. The user agent MAY make a selection from that list + automatically if it understands the provided media type. A specific format + for automatic selection is not defined by this specification because HTTP + tries to remain orthogonal to the definition of its content. + In practice, the representation is provided in some easily parsed format + believed to be acceptable to the user agent, as determined by shared design + or content negotiation, or in some commonly accepted hypertext format.¶

+

+ A 300 response is heuristically cacheable; i.e., unless otherwise indicated by + the method definition or explicit cache controls (see Section 4.2.2 of [CACHING]).¶

+ +
+
+
+
+

+15.4.2. 301 Moved Permanently +

+ +

+ The 301 (Moved Permanently) status code indicates that the + target resource has been assigned a new permanent URI and + any future references to this resource ought to use one of the enclosed + URIs. The server is suggesting that a user agent with link-editing capability + can permanently replace references to the target URI with one of the + new references sent by the server. However, this suggestion is usually + ignored unless the user agent is actively editing references + (e.g., engaged in authoring content), the connection is secured, and + the origin server is a trusted authority for the content being edited.¶

+

+ The server SHOULD generate a Location header field in the + response containing a preferred URI reference for the new permanent URI. + The user agent MAY use the Location field value for automatic redirection. + The server's response content usually contains a short hypertext note with + a hyperlink to the new URI(s).¶

+ +

+ A 301 response is heuristically cacheable; i.e., unless otherwise indicated by + the method definition or explicit cache controls (see Section 4.2.2 of [CACHING]).¶

+
+
+
+
+

+15.4.3. 302 Found +

+ +

+ The 302 (Found) status code indicates that the target + resource resides temporarily under a different URI. Since the redirection + might be altered on occasion, the client ought to continue to use the + target URI for future requests.¶

+

+ The server SHOULD generate a Location header field in the + response containing a URI reference for the different URI. + The user agent MAY use the Location field value for automatic redirection. + The server's response content usually contains a short hypertext note with + a hyperlink to the different URI(s).¶

+ +
+
+
+
+

+15.4.4. 303 See Other +

+ +

+ The 303 (See Other) status code indicates that the server is + redirecting the user agent to a different resource, as indicated by a URI + in the Location header field, which is intended to provide + an indirect response to the original request. A user agent can perform a + retrieval request targeting that URI (a GET or HEAD request if using HTTP), + which might also be redirected, and present the eventual result as an + answer to the original request. Note that the new URI in the Location + header field is not considered equivalent to the target URI.¶

+

+ This status code is applicable to any HTTP method. It is + primarily used to allow the output of a POST action to redirect + the user agent to a different resource, since doing so provides the + information corresponding to the POST response as a resource that + can be separately identified, bookmarked, and cached.¶

+

+ A 303 response to a GET request indicates that the origin server does not + have a representation of the target resource that can be + transferred by the server over HTTP. However, the + Location field value refers to a resource that is + descriptive of the target resource, such that making a retrieval request + on that other resource might result in a representation that is useful to + recipients without implying that it represents the original target resource. + Note that answers to the questions of what can be represented, what + representations are adequate, and what might be a useful description are + outside the scope of HTTP.¶

+

+ Except for responses to a HEAD request, the representation of a 303 + response ought to contain a short hypertext note with a hyperlink to the + same URI reference provided in the Location header field.¶

+
+
+
+
+

+15.4.5. 304 Not Modified +

+ +

+ The 304 (Not Modified) status code indicates that a + conditional GET or HEAD request has been + received and would have resulted in a 200 (OK) response + if it were not for the fact that the condition evaluated to false. + In other words, there is no need for the server to transfer a + representation of the target resource because the request indicates that + the client, which made the request conditional, already has a valid + representation; the server is therefore redirecting the client to make + use of that stored representation as if it were the content of a + 200 (OK) response.¶

+

+ The server generating a 304 response MUST generate any of the following + header fields that would have been sent in a 200 (OK) + response to the same request:¶

+ +

+ Since the goal of a 304 response is to minimize information transfer + when the recipient already has one or more cached representations, + a sender SHOULD NOT generate representation metadata other + than the above listed fields unless said metadata exists for the + purpose of guiding cache updates (e.g., Last-Modified might + be useful if the response does not have an ETag field).¶

+

+ Requirements on a cache that receives a 304 response are defined in + Section 4.3.4 of [CACHING]. If the conditional request originated with an + outbound client, such as a user agent with its own cache sending a + conditional GET to a shared proxy, then the proxy SHOULD forward the + 304 response to that client.¶

+

+ A 304 response is terminated by the end of the header section; + it cannot contain content or trailers.¶

+
+
+
+
+

+15.4.6. 305 Use Proxy +

+ +

+ The 305 (Use Proxy) status code was defined in a previous + version of this specification and is now deprecated (Appendix B of [RFC7231]).¶

+
+
+
+
+

+15.4.7. 306 (Unused) +

+ +

+ The 306 status code was defined in a previous version of this + specification, is no longer used, and the code is reserved.¶

+
+
+
+
+

+15.4.8. 307 Temporary Redirect +

+ +

+ The 307 (Temporary Redirect) status code indicates that the + target resource resides temporarily under a different URI + and the user agent MUST NOT change the request method if it performs an + automatic redirection to that URI. + Since the redirection can change over time, the client ought to continue + using the original target URI for future requests.¶

+

+ The server SHOULD generate a Location header field in the + response containing a URI reference for the different URI. + The user agent MAY use the Location field value for automatic redirection. + The server's response content usually contains a short hypertext note with + a hyperlink to the different URI(s).¶

+
+
+
+
+

+15.4.9. 308 Permanent Redirect +

+ +

+ The 308 (Permanent Redirect) status code indicates that the + target resource has been assigned a new permanent URI and + any future references to this resource ought to use one of the enclosed + URIs. The server is suggesting that a user agent with link-editing capability + can permanently replace references to the target URI with one of the + new references sent by the server. However, this suggestion is usually + ignored unless the user agent is actively editing references + (e.g., engaged in authoring content), the connection is secured, and + the origin server is a trusted authority for the content being edited.¶

+

+ The server SHOULD generate a Location header field in the + response containing a preferred URI reference for the new permanent URI. + The user agent MAY use the Location field value for automatic redirection. + The server's response content usually contains a short hypertext note with + a hyperlink to the new URI(s).¶

+

+ A 308 response is heuristically cacheable; i.e., unless otherwise indicated by + the method definition or explicit cache controls (see Section 4.2.2 of [CACHING]).¶

+ +
+
+
+
+
+
+

+15.5. Client Error 4xx +

+ + +

+ The 4xx (Client Error) class of status code indicates that + the client seems to have erred. Except when responding to a HEAD request, + the server SHOULD send a representation containing an explanation of + the error situation, and whether it is a temporary or permanent condition. + These status codes are applicable to any request method. User agents + SHOULD display any included representation to the user.¶

+
+
+

+15.5.1. 400 Bad Request +

+ +

+ The 400 (Bad Request) status code indicates that the server + cannot or will not process the request due to something that is perceived + to be a client error (e.g., malformed request syntax, invalid request + message framing, or deceptive request routing).¶

+
+
+
+
+

+15.5.2. 401 Unauthorized +

+ +

+ The 401 (Unauthorized) status code indicates that the + request has not been applied because it lacks valid authentication + credentials for the target resource. + The server generating a 401 response MUST send a + WWW-Authenticate header field + (Section 11.6.1) + containing at least one challenge applicable to the target resource.¶

+

+ If the request included authentication credentials, then the 401 response + indicates that authorization has been refused for those credentials. + The user agent MAY repeat the request with a new or replaced + Authorization header field (Section 11.6.2). + If the 401 response contains the same challenge as the prior response, and + the user agent has already attempted authentication at least once, then the + user agent SHOULD present the enclosed representation to the user, since + it usually contains relevant diagnostic information.¶

+
+
+
+
+

+15.5.3. 402 Payment Required +

+ +

+ The 402 (Payment Required) status code is reserved for + future use.¶

+
+
+
+
+

+15.5.4. 403 Forbidden +

+ +

+ The 403 (Forbidden) status code indicates that the + server understood the request but refuses to fulfill it. + A server that wishes to make public why the request has been forbidden + can describe that reason in the response content (if any).¶

+

+ If authentication credentials were provided in the request, the + server considers them insufficient to grant access. + The client SHOULD NOT automatically repeat the request with the same + credentials. + The client MAY repeat the request with new or different credentials. + However, a request might be forbidden for reasons unrelated to the + credentials.¶

+

+ An origin server that wishes to "hide" the current existence of a forbidden + target resource + MAY instead respond with a status + code of 404 (Not Found).¶

+
+
+
+
+

+15.5.5. 404 Not Found +

+ +

+ The 404 (Not Found) status code indicates that the origin + server did not find a current representation for the + target resource or is not willing to disclose that one + exists. A 404 status code does not indicate whether this lack of representation + is temporary or permanent; the 410 (Gone) status code is + preferred over 404 if the origin server knows, presumably through some + configurable means, that the condition is likely to be permanent.¶

+

+ A 404 response is heuristically cacheable; i.e., unless otherwise indicated by + the method definition or explicit cache controls (see Section 4.2.2 of [CACHING]).¶

+
+
+
+
+

+15.5.6. 405 Method Not Allowed +

+ +

+ The 405 (Method Not Allowed) status code indicates that the + method received in the request-line is known by the origin server but + not supported by the target resource. + The origin server MUST generate an Allow header field in + a 405 response containing a list of the target resource's currently + supported methods.¶

+

+ A 405 response is heuristically cacheable; i.e., unless otherwise indicated by + the method definition or explicit cache controls (see Section 4.2.2 of [CACHING]).¶

+
+
+
+
+

+15.5.7. 406 Not Acceptable +

+ +

+ The 406 (Not Acceptable) status code indicates that the + target resource does not have a current representation that + would be acceptable to the user agent, according to the + proactive negotiation header fields received in the request + (Section 12.1), and the server is unwilling to supply a + default representation.¶

+

+ The server SHOULD generate content containing a list of available + representation characteristics and corresponding resource identifiers from + which the user or user agent can choose the one most appropriate. + A user agent MAY automatically select the most appropriate choice from + that list. However, this specification does not define any standard for + such automatic selection, as described in Section 15.4.1.¶

+
+
+
+
+

+15.5.8. 407 Proxy Authentication Required +

+ +

+ The 407 (Proxy Authentication Required) status code is + similar to 401 (Unauthorized), but it indicates that the client + needs to authenticate itself in order to use a proxy for this request. + The proxy MUST send a Proxy-Authenticate header field + (Section 11.7.1) containing a challenge + applicable to that proxy for the request. The client MAY repeat + the request with a new or replaced Proxy-Authorization + header field (Section 11.7.2).¶

+
+
+
+
+

+15.5.9. 408 Request Timeout +

+ +

+ The 408 (Request Timeout) status code indicates + that the server did not receive a complete request message within the time + that it was prepared to wait.¶

+

+ If the client has an outstanding request in transit, it MAY repeat that + request. If the current connection is not usable (e.g., as it would be in + HTTP/1.1 because request delimitation is lost), a new connection will be + used.¶

+
+
+
+
+

+15.5.10. 409 Conflict +

+ +

+ The 409 (Conflict) status code indicates that the request + could not be completed due to a conflict with the current state of the target + resource. This code is used in situations where the user might be able to + resolve the conflict and resubmit the request. The server SHOULD generate + content that includes enough information for a user to recognize the + source of the conflict.¶

+

+ Conflicts are most likely to occur in response to a PUT request. For + example, if versioning were being used and the representation being PUT + included changes to a resource that conflict with those made by an + earlier (third-party) request, the origin server might use a 409 response + to indicate that it can't complete the request. In this case, the response + representation would likely contain information useful for merging the + differences based on the revision history.¶

+
+
+
+
+

+15.5.11. 410 Gone +

+ +

+ The 410 (Gone) status code indicates that access to the + target resource is no longer available at the origin + server and that this condition is likely to be permanent. If the origin + server does not know, or has no facility to determine, whether or not the + condition is permanent, the status code 404 (Not Found) + ought to be used instead.¶

+

+ The 410 response is primarily intended to assist the task of web + maintenance by notifying the recipient that the resource is + intentionally unavailable and that the server owners desire that + remote links to that resource be removed. Such an event is common for + limited-time, promotional services and for resources belonging to + individuals no longer associated with the origin server's site. It is not + necessary to mark all permanently unavailable resources as "gone" or + to keep the mark for any length of time -- that is left to the + discretion of the server owner.¶

+

+ A 410 response is heuristically cacheable; i.e., unless otherwise indicated by + the method definition or explicit cache controls (see Section 4.2.2 of [CACHING]).¶

+
+
+
+
+

+15.5.12. 411 Length Required +

+ +

+ The 411 (Length Required) status code indicates that the + server refuses to accept the request without a defined + Content-Length (Section 8.6). + The client MAY repeat the request if it adds a valid Content-Length + header field containing the length of the request content.¶

+
+
+
+
+

+15.5.13. 412 Precondition Failed +

+ +

+ The 412 (Precondition Failed) status code indicates that one + or more conditions given in the request header fields evaluated to false + when tested on the server (Section 13). This + response status code allows the client to place preconditions on the + current resource state (its current representations and metadata) and, + thus, prevent the request method from being applied if the target resource + is in an unexpected state.¶

+
+
+
+
+

+15.5.14. 413 Content Too Large +

+ +

+ The 413 (Content Too Large) status code indicates + that the server is refusing to process a request because the request + content is larger than the server is willing or able to process. + The server MAY terminate the request, if the protocol version in use + allows it; otherwise, the server MAY close the connection.¶

+

+ If the condition is temporary, the server SHOULD generate a + Retry-After header field to indicate that it is temporary + and after what time the client MAY try again.¶

+
+
+
+
+

+15.5.15. 414 URI Too Long +

+ +

+ The 414 (URI Too Long) status code indicates that the server + is refusing to service the request because the + target URI is longer than the server is willing to + interpret. This rare condition is only likely to occur when a client has + improperly converted a POST request to a GET request with long query + information, when the client has descended into an infinite loop of + redirection (e.g., a redirected URI prefix that points to a suffix of + itself) or when the server is under attack by a client attempting to + exploit potential security holes.¶

+

+ A 414 response is heuristically cacheable; i.e., unless otherwise indicated by + the method definition or explicit cache controls (see Section 4.2.2 of [CACHING]).¶

+
+
+
+
+

+15.5.16. 415 Unsupported Media Type +

+ +

+ The 415 (Unsupported Media Type) status code indicates that + the origin server is refusing to service the request because the content is + in a format not supported by this method on the target resource.¶

+

+ The format problem might be due to the request's indicated + Content-Type or Content-Encoding, or as a + result of inspecting the data directly.¶

+

+ If the problem was caused by an unsupported content coding, the + Accept-Encoding response header field + (Section 12.5.3) ought to be + used to indicate which (if any) content codings would have been accepted + in the request.¶

+

+ On the other hand, if the cause was an unsupported media type, the + Accept response header field (Section 12.5.1) + can be used to indicate which media types would have been accepted + in the request.¶

+
+
+
+
+

+15.5.17. 416 Range Not Satisfiable +

+ +

+ The 416 (Range Not Satisfiable) status code indicates that + the set of ranges in the request's Range header field + (Section 14.2) has been rejected either because none of + the requested ranges are satisfiable or because the client has requested + an excessive number of small or overlapping ranges (a potential denial of + service attack).¶

+

+ Each range unit defines what is required for its own range sets to be + satisfiable. For example, Section 14.1.2 defines what makes + a bytes range set satisfiable.¶

+

+ A server that generates a 416 response to a byte-range request SHOULD + generate a Content-Range header field + specifying the current length of the selected representation + (Section 14.4).¶

+

+ For example:¶

+
+
HTTP/1.1 416 Range Not Satisfiable
+Date: Fri, 20 Jan 2012 15:41:54 GMT
+Content-Range: bytes */47022
+
¶ +
+ +
+
+
+
+

+15.5.18. 417 Expectation Failed +

+ +

+ The 417 (Expectation Failed) status code indicates that the + expectation given in the request's Expect header field + (Section 10.1.1) could not be met by at least one of the + inbound servers.¶

+
+
+
+
+

+15.5.19. 418 (Unused) +

+ +

+ [RFC2324] was an April 1 RFC that lampooned the various ways + HTTP was abused; one such abuse was the definition of an + application-specific 418 status code, which has been deployed as a joke + often enough for the code to be unusable for any future use.¶

+

+ Therefore, the 418 status code is reserved in the IANA HTTP Status Code + Registry. This indicates that the status code cannot be assigned to other + applications currently. If future circumstances require its use (e.g., + exhaustion of 4NN status codes), it can be re-assigned to another use.¶

+
+
+
+
+

+15.5.20. 421 Misdirected Request +

+ +

+ The 421 (Misdirected Request) status code indicates that the request was + directed at a server that is unable or unwilling to produce an + authoritative response for the target URI. An origin server (or gateway + acting on behalf of the origin server) sends 421 to reject a target URI + that does not match an origin for which the server has been + configured (Section 4.3.1) or does not match the connection + context over which the request was received + (Section 7.4).¶

+

+ A client that receives a 421 (Misdirected Request) response MAY retry the + request, whether or not the request method is idempotent, over a different + connection, such as a fresh connection specific to the target resource's + origin, or via an alternative service [ALTSVC].¶

+

+ A proxy MUST NOT generate a 421 response.¶

+
+
+
+
+

+15.5.21. 422 Unprocessable Content +

+ +

+ The 422 (Unprocessable Content) status code indicates that the server + understands the content type of the request content (hence a + 415 (Unsupported Media Type) status code is inappropriate), + and the syntax of the request content is correct, but it was unable to process + the contained instructions. For example, this status code can be sent if + an XML request content contains well-formed (i.e., syntactically correct), but + semantically erroneous XML instructions.¶

+
+
+
+
+

+15.5.22. 426 Upgrade Required +

+ +

+ The 426 (Upgrade Required) status code indicates that the + server refuses to perform the request using the current protocol but might + be willing to do so after the client upgrades to a different protocol. + The server MUST send an Upgrade header field in a 426 + response to indicate the required protocol(s) (Section 7.8).¶

+

+ Example:¶

+
+
HTTP/1.1 426 Upgrade Required
+Upgrade: HTTP/3.0
+Connection: Upgrade
+Content-Length: 53
+Content-Type: text/plain
+
+This service requires use of the HTTP/3.0 protocol.
+
¶ +
+
+
+
+
+
+
+

+15.6. Server Error 5xx +

+ + +

+ The 5xx (Server Error) class of status code indicates that + the server is aware that it has erred or is incapable of performing the + requested method. + Except when responding to a HEAD request, the server SHOULD send a + representation containing an explanation of the error situation, and + whether it is a temporary or permanent condition. + A user agent SHOULD display any included representation to the user. + These status codes are applicable to any request method.¶

+
+
+

+15.6.1. 500 Internal Server Error +

+ +

+ The 500 (Internal Server Error) status code indicates that + the server encountered an unexpected condition that prevented it from + fulfilling the request.¶

+
+
+
+
+

+15.6.2. 501 Not Implemented +

+ +

+ The 501 (Not Implemented) status code indicates that the + server does not support the functionality required to fulfill the request. + This is the appropriate response when the server does not recognize the + request method and is not capable of supporting it for any resource.¶

+

+ A 501 response is heuristically cacheable; i.e., unless otherwise indicated by + the method definition or explicit cache controls (see Section 4.2.2 of [CACHING]).¶

+
+
+
+
+

+15.6.3. 502 Bad Gateway +

+ +

+ The 502 (Bad Gateway) status code indicates that the server, + while acting as a gateway or proxy, received an invalid response from an + inbound server it accessed while attempting to fulfill the request.¶

+
+
+
+
+

+15.6.4. 503 Service Unavailable +

+ +

+ The 503 (Service Unavailable) status code indicates that the + server is currently unable to handle the request due to a temporary overload + or scheduled maintenance, which will likely be alleviated after some delay. + The server MAY send a Retry-After header field + (Section 10.2.3) to suggest an appropriate + amount of time for the client to wait before retrying the request.¶

+ +
+
+
+
+

+15.6.5. 504 Gateway Timeout +

+ +

+ The 504 (Gateway Timeout) status code indicates that the + server, while acting as a gateway or proxy, did not receive a timely + response from an upstream server it needed to access in order to + complete the request.¶

+
+
+
+
+

+15.6.6. 505 HTTP Version Not Supported +

+ +

+ The 505 (HTTP Version Not Supported) status code indicates + that the server does not support, or refuses to support, the major version + of HTTP that was used in the request message. The server is indicating that + it is unable or unwilling to complete the request using the same major + version as the client, as described in Section 2.5, other than with this + error message. The server SHOULD generate a representation for the 505 + response that describes why that version is not supported and what other + protocols are supported by that server.¶

+
+
+
+
+
+
+
+
+

+16. Extending HTTP +

+

+ HTTP defines a number of generic extension points that can be used to + introduce capabilities to the protocol without introducing a new version, + including methods, status codes, field names, and further extensibility + points within defined fields, such as authentication schemes and + cache directives (see Cache-Control extensions in Section 5.2.3 of [CACHING]). Because the semantics of HTTP are + not versioned, these extension points are persistent; the version of the + protocol in use does not affect their semantics.¶

+

+ Version-independent extensions are discouraged from depending on or + interacting with the specific version of the protocol in use. When this is + unavoidable, careful consideration needs to be given to how the extension + can interoperate across versions.¶

+

+ Additionally, specific versions of HTTP might have their own extensibility + points, such as transfer codings in HTTP/1.1 (Section 6.1 of [HTTP/1.1]) and HTTP/2 SETTINGS or frame types + ([HTTP/2]). These extension points are specific to the + version of the protocol they occur within.¶

+

+ Version-specific extensions cannot override or modify the semantics of + a version-independent mechanism or extension point (like a method or + header field) without explicitly being allowed by that protocol element. For + example, the CONNECT method (Section 9.3.6) allows this.¶

+

+ These guidelines assure that the protocol operates correctly and + predictably, even when parts of the path implement different versions of + HTTP.¶

+
+
+

+16.1. Method Extensibility +

+
+
+

+16.1.1. Method Registry +

+

+ The "Hypertext Transfer Protocol (HTTP) Method Registry", maintained by + IANA at <https://www.iana.org/assignments/http-methods>, + registers method names.¶

+

+ HTTP method registrations MUST include the following fields:¶

+ +

+ Values to be added to this namespace require IETF Review + (see [RFC8126], Section 4.8).¶

+
+
+
+
+

+16.1.2. Considerations for New Methods +

+

+ Standardized methods are generic; that is, they are potentially + applicable to any resource, not just one particular media type, kind of + resource, or application. As such, it is preferred that new methods + be registered in a document that isn't specific to a single application or + data format, since orthogonal technologies deserve orthogonal specification.¶

+

+ Since message parsing (Section 6) needs to be + independent of method + semantics (aside from responses to HEAD), definitions of new methods + cannot change the parsing algorithm or prohibit the presence of content + on either the request or the response message. + Definitions of new methods can specify that only a zero-length content + is allowed by requiring a Content-Length header field with a value of "0".¶

+

+ Likewise, new methods cannot use the special host:port and asterisk forms of + request target that are allowed for CONNECT and + OPTIONS, respectively (Section 7.1). + A full URI in absolute form is needed for the target URI, which means either + the request target needs to be sent in absolute form or the target URI will + be reconstructed from the request context in the same way it is for other + methods.¶

+

+ A new method definition needs to indicate whether it is safe (Section 9.2.1), idempotent (Section 9.2.2), + cacheable (Section 9.2.3), what + semantics are to be associated with the request content (if any), and what + refinements the method makes to header field or status code semantics. + If the new method is cacheable, its definition ought to describe how, and + under what conditions, a cache can store a response and use it to satisfy a + subsequent request. + The new method ought to describe whether it can be made conditional + (Section 13.1) and, if so, how a server responds + when the condition is false. + Likewise, if the new method might have some use for partial response + semantics (Section 14.2), it ought to document this, too.¶

+ +
+
+
+
+
+
+

+16.2. Status Code Extensibility +

+
+
+

+16.2.1. Status Code Registry +

+

+ The "Hypertext Transfer Protocol (HTTP) Status Code Registry", maintained + by IANA at <https://www.iana.org/assignments/http-status-codes>, + registers status code numbers.¶

+

+ A registration MUST include the following fields:¶

+
    +
  • Status Code (3 digits)¶ +
    • +
  • Short Description¶ +
    • +
  • Pointer to specification text¶ +
    • +
+

+ Values to be added to the HTTP status code namespace require IETF Review + (see [RFC8126], Section 4.8).¶

+
+
+
+
+

+16.2.2. Considerations for New Status Codes +

+

+ When it is necessary to express semantics for a response that are not + defined by current status codes, a new status code can be registered. + Status codes are generic; they are potentially applicable to any resource, + not just one particular media type, kind of resource, or application of + HTTP. As such, it is preferred that new status codes be registered in a + document that isn't specific to a single application.¶

+

+ New status codes are required to fall under one of the categories + defined in Section 15. To allow existing parsers to + process the response message, new status codes cannot disallow content, + although they can mandate a zero-length content.¶

+

+ Proposals for new status codes that are not yet widely deployed ought to + avoid allocating a specific number for the code until there is clear + consensus that it will be registered; instead, early drafts can use a + notation such as "4NN", or "3N0" .. "3N9", to indicate the class + of the proposed status code(s) without consuming a number prematurely.¶

+

+ The definition of a new status code ought to explain the request + conditions that would cause a response containing that status code (e.g., + combinations of request header fields and/or method(s)) along with any + dependencies on response header fields (e.g., what fields are required, + what fields can modify the semantics, and what field semantics are + further refined when used with the new status code).¶

+

+ By default, a status code applies only to the request corresponding to the + response it occurs within. If a status code applies to a larger scope of + applicability -- for example, all requests to the resource in question or + all requests to a server -- this must be explicitly specified. When doing + so, it should be noted that not all clients can be expected to + consistently apply a larger scope because they might not understand the + new status code.¶

+

+ The definition of a new final status code ought to specify whether or not it + is heuristically cacheable. Note that any response with a final status code + can be cached if the response has explicit freshness information. A status + code defined as heuristically cacheable is allowed to be cached without + explicit freshness information. + Likewise, the definition of a status code can place + constraints upon cache behavior if the must-understand cache + directive is used. See [CACHING] for more information.¶

+

+ Finally, the definition of a new status code ought to indicate whether the + content has any implied association with an identified resource (Section 6.4.2).¶

+
+
+
+
+
+
+

+16.3. Field Extensibility +

+

+ HTTP's most widely used extensibility point is the definition of new header and + trailer fields.¶

+

+ New fields can be defined such that, when they are understood by a + recipient, they override or enhance the interpretation of previously + defined fields, define preconditions on request evaluation, or + refine the meaning of responses.¶

+

+ However, defining a field doesn't guarantee its deployment or recognition + by recipients. Most fields are designed with the expectation that a recipient + can safely ignore (but forward downstream) any field not recognized. + In other cases, the sender's ability to understand a given field might be + indicated by its prior communication, perhaps in the protocol version + or fields that it sent in prior messages, or its use of a specific media type. + Likewise, direct inspection of support might be possible through an + OPTIONS request or by interacting with a defined well-known URI + [RFC8615] if such inspection is defined along with + the field being introduced.¶

+
+
+

+16.3.1. Field Name Registry +

+

+ The "Hypertext Transfer Protocol (HTTP) Field Name Registry" defines the + namespace for HTTP field names.¶

+

+ Any party can request registration of an HTTP field. See Section 16.3.2 for considerations to take + into account when creating a new HTTP field.¶

+

+ The "Hypertext Transfer Protocol (HTTP) Field Name Registry" is located at + <https://www.iana.org/assignments/http-fields/>. + Registration requests can be made by following the instructions located + there or by sending an email to the "ietf-http-wg@w3.org" mailing list.¶

+

+ Field names are registered on the advice of a designated expert + (appointed by the IESG or their delegate). Fields with the status + 'permanent' are Specification Required + ([RFC8126], Section 4.6).¶

+

+ Registration requests consist of the following information:¶

+
+
Field name:
+
+ The requested field name. It MUST conform to the + field-name syntax defined in Section 5.1, and it SHOULD be + restricted to just letters, digits, and hyphen ('-') + characters, with the first character being a letter.¶ +
+
+
Status:
+
+ "permanent", "provisional", "deprecated", or "obsoleted".¶ +
+
+
Specification document(s):
+
+ Reference to the document that specifies + the field, preferably including a URI that can be used to retrieve + a copy of the document. Optional but encouraged for provisional registrations. + An indication of the relevant section(s) can also be included, but is not required.¶ +
+
+
+

+ And optionally:¶

+
+
Comments:
+
+ Additional information, such as about reserved entries.¶ +
+
+
+

+ The expert(s) can define additional fields to be collected in the + registry, in consultation with the community.¶

+

+ Standards-defined names have a status of "permanent". Other names can also + be registered as permanent if the expert(s) finds that they are in use, in + consultation with the community. Other names should be registered as + "provisional".¶

+

+ Provisional entries can be removed by the expert(s) if -- in consultation + with the community -- the expert(s) find that they are not in use. The + expert(s) can change a provisional entry's status to permanent at any time.¶

+

+ Note that names can be registered by third parties (including the + expert(s)) if the expert(s) determines that an unregistered name is widely + deployed and not likely to be registered in a timely manner otherwise.¶

+
+
+
+
+

+16.3.2. Considerations for New Fields +

+

+ HTTP header and trailer fields are a widely used extension point for the protocol. + While they can be used in an ad hoc fashion, fields that are intended for + wider use need to be carefully documented to ensure interoperability.¶

+

+ In particular, authors of specifications defining new fields are advised to consider + and, where appropriate, document the following aspects:¶

+
    +
  • Under what conditions the field can be used; e.g., only in + responses or requests, in all messages, only on responses to a + particular request method, etc.¶ +
    • +
  • Whether the field semantics are further refined by their context, + such as their use with certain request methods or status codes.¶ +
    • +
  • The scope of applicability for the information conveyed. + By default, fields apply only to the message they are + associated with, but some response fields are designed to apply to all + representations of a resource, the resource itself, or an even broader + scope. Specifications that expand the scope of a response field will + need to carefully consider issues such as content negotiation, the time + period of applicability, and (in some cases) multi-tenant server + deployments.¶ +
    • +
  • Under what conditions intermediaries are allowed to insert, + delete, or modify the field's value.¶ +
    • +
  • If the field is allowable in trailers; by + default, it will not be (see Section 6.5.1).¶ +
    • +
  • Whether it is appropriate or even required to list the field name in the + Connection header field (i.e., if the field is to + be hop-by-hop; see Section 7.6.1).¶ +
    • +
  • Whether the field introduces any additional security considerations, such + as disclosure of privacy-related data.¶ +
    • +
+

+ Request header fields have additional considerations that need to be documented + if the default behavior is not appropriate:¶

+
    +
  • If it is appropriate to list the field name in a + Vary response header field (e.g., when the request header + field is used by an origin server's content selection algorithm; see + Section 12.5.5).¶ +
    • +
  • If the field is intended to be stored when received in a PUT + request (see Section 9.3.4).¶ +
    • +
  • If the field ought to be removed when automatically redirecting a + request due to security concerns (see Section 15.4).¶ +
    • +
+
+
+
+16.3.2.1. Considerations for New Field Names +
+

+ Authors of specifications defining new fields are advised to choose a short + but descriptive field name. Short names avoid needless data transmission; + descriptive names avoid confusion and "squatting" on names that might have + broader uses.¶

+

+ To that end, limited-use fields (such as a header confined to a single + application or use case) are encouraged to use a name that includes that use + (or an abbreviation) as a prefix; for example, if the Foo Application needs + a Description field, it might use "Foo-Desc"; "Description" is too generic, + and "Foo-Description" is needlessly long.¶

+

+ While the field-name syntax is defined to allow any token character, in + practice some implementations place limits on the characters they accept + in field-names. To be interoperable, new field names SHOULD constrain + themselves to alphanumeric characters, "-", and ".", and SHOULD + begin with a letter. For example, the underscore + ("_") character can be problematic when passed through non-HTTP + gateway interfaces (see Section 17.10).¶

+

+ Field names ought not be prefixed with "X-"; see + [BCP178] for further information.¶

+

+ Other prefixes are sometimes used in HTTP field names; for example, + "Accept-" is used in many content negotiation headers, and "Content-" is used + as explained in Section 6.4. These prefixes are + only an aid to recognizing the purpose of a field and do not + trigger automatic processing.¶

+
+
+
+
+
+16.3.2.2. Considerations for New Field Values +
+

+ A major task in the definition of a new HTTP field is the specification of + the field value syntax: what senders should generate, and how recipients + should infer semantics from what is received.¶

+

+ Authors are encouraged (but not required) to use either the ABNF rules in + this specification or those in [RFC8941] to define the syntax + of new field values.¶

+

+ Authors are advised to carefully consider how the combination of multiple + field lines will impact them (see Section 5.3). Because + senders might erroneously send multiple values, and both intermediaries + and HTTP libraries can perform combination automatically, this applies to + all field values -- even when only a single value is anticipated.¶

+

+ Therefore, authors are advised to delimit or encode values that contain + commas (e.g., with the quoted-string rule of + Section 5.6.4, the String data type of + [RFC8941], or a field-specific encoding). + This ensures that commas within field data are not confused + with the commas that delimit a list value.¶

+

+ For example, the Content-Type field value only allows commas + inside quoted strings, which can be reliably parsed even when multiple + values are present. The Location field value provides a + counter-example that should not be emulated: because URIs can include + commas, it is not possible to reliably distinguish between a single value + that includes a comma from two values.¶

+

+ Authors of fields with a singleton value (see Section 5.5) are additionally advised to document how to treat + messages where the multiple members are present (a sensible default would + be to ignore the field, but this might not always be the right choice).¶

+
+
+
+
+
+
+
+
+

+16.4. Authentication Scheme Extensibility +

+
+
+

+16.4.1. Authentication Scheme Registry +

+

+ The "Hypertext Transfer Protocol (HTTP) Authentication Scheme Registry" + defines the namespace for the authentication schemes in challenges and + credentials. It is maintained + at <https://www.iana.org/assignments/http-authschemes>.¶

+

+ Registrations MUST include the following fields:¶

+
    +
  • Authentication Scheme Name¶ +
    • +
  • Pointer to specification text¶ +
    • +
  • Notes (optional)¶ +
    • +
+

+ Values to be added to this namespace require IETF Review + (see [RFC8126], Section 4.8).¶

+
+
+
+
+

+16.4.2. Considerations for New Authentication Schemes +

+

+ There are certain aspects of the HTTP Authentication framework that put + constraints on how new authentication schemes can work:¶

+
    +
  • +

    + HTTP authentication is presumed to be stateless: all of the information + necessary to authenticate a request MUST be provided in the request, + rather than be dependent on the server remembering prior requests. + Authentication based on, or bound to, the underlying connection is + outside the scope of this specification and inherently flawed unless + steps are taken to ensure that the connection cannot be used by any + party other than the authenticated user + (see Section 3.3).¶

    +
    • +
  • +

    + The authentication parameter "realm" is reserved for defining protection + spaces as described in Section 11.5. New schemes + MUST NOT use it in a way incompatible with that definition.¶

    +
    • +
  • +

    + The "token68" notation was introduced for compatibility with existing + authentication schemes and can only be used once per challenge or credential. + Thus, new schemes ought to use the auth-param syntax instead, because + otherwise future extensions will be impossible.¶

    +
    • +
  • +

    + The parsing of challenges and credentials is defined by this specification + and cannot be modified by new authentication schemes. When the auth-param + syntax is used, all parameters ought to support both token and + quoted-string syntax, and syntactical constraints ought to be defined on + the field value after parsing (i.e., quoted-string processing). This is + necessary so that recipients can use a generic parser that applies to + all authentication schemes.¶

    +

    + Note: The fact that the value syntax for the "realm" parameter + is restricted to quoted-string was a bad design choice not to be repeated + for new parameters.¶

    +
    • +
  • +

    + Definitions of new schemes ought to define the treatment of unknown + extension parameters. In general, a "must-ignore" rule is preferable + to a "must-understand" rule, because otherwise it will be hard to introduce + new parameters in the presence of legacy recipients. Furthermore, + it's good to describe the policy for defining new parameters (such + as "update the specification" or "use this registry").¶

    +
    • +
  • +

    + Authentication schemes need to document whether they are usable in + origin-server authentication (i.e., using WWW-Authenticate), + and/or proxy authentication (i.e., using Proxy-Authenticate).¶

    +
    • +
  • +

    + The credentials carried in an Authorization header field are specific to + the user agent and, therefore, have the same effect on HTTP caches as the + "private" cache response directive (Section 5.2.2.7 of [CACHING]), + within the scope of the request in which they appear.¶

    +

    + Therefore, new authentication schemes that choose not to carry + credentials in the Authorization header field (e.g., using a newly defined + header field) will need to explicitly disallow caching, by mandating the use of + cache response directives (e.g., "private").¶

    +
    • +
  • +

    + Schemes using Authentication-Info, Proxy-Authentication-Info, + or any other authentication related response header field need to + consider and document the related security considerations (see + Section 17.16.4).¶

    +
    • +
+
+
+
+
+
+
+

+16.5. Range Unit Extensibility +

+
+
+

+16.5.1. Range Unit Registry +

+

+ The "HTTP Range Unit Registry" defines the namespace for the range + unit names and refers to their corresponding specifications. + It is maintained at + <https://www.iana.org/assignments/http-parameters>.¶

+

+ Registration of an HTTP Range Unit MUST include the following fields:¶

+
    +
  • Name¶ +
    • +
  • Description¶ +
    • +
  • Pointer to specification text¶ +
    • +
+

+ Values to be added to this namespace require IETF Review + (see [RFC8126], Section 4.8).¶

+
+
+
+
+

+16.5.2. Considerations for New Range Units +

+

+ Other range units, such as format-specific boundaries like pages, + sections, records, rows, or time, are potentially usable in HTTP for + application-specific purposes, but are not commonly used in practice. + Implementors of alternative range units ought to consider how they would + work with content codings and general-purpose intermediaries.¶

+
+
+
+
+
+
+

+16.6. Content Coding Extensibility +

+
+
+

+16.6.1. Content Coding Registry +

+

+ The "HTTP Content Coding Registry", maintained by + IANA at <https://www.iana.org/assignments/http-parameters/>, + registers content-coding names.¶

+

+ Content coding registrations MUST include the following fields:¶

+
    +
  • Name¶ +
    • +
  • Description¶ +
    • +
  • Pointer to specification text¶ +
    • +
+

+ Names of content codings MUST NOT overlap with names of transfer codings + (per the "HTTP Transfer Coding Registry" located at + <https://www.iana.org/assignments/http-parameters/>) unless + the encoding transformation is + identical (as is the case for the compression codings defined in + Section 8.4.1).¶

+

+ Values to be added to this namespace require IETF Review + (see Section 4.8 of [RFC8126]) and MUST + conform to the purpose of content coding defined in + Section 8.4.1.¶

+
+
+
+
+

+16.6.2. Considerations for New Content Codings +

+

+ New content codings ought to be self-descriptive whenever possible, with + optional parameters discoverable within the coding format itself, rather + than rely on external metadata that might be lost during transit.¶

+
+
+
+
+
+
+

+16.7. Upgrade Token Registry +

+

+ The "Hypertext Transfer Protocol (HTTP) Upgrade Token Registry" defines + the namespace for protocol-name tokens used to identify protocols in the + Upgrade header field. The registry is maintained at + <https://www.iana.org/assignments/http-upgrade-tokens>.¶

+

+ Each registered protocol name is associated with contact information + and an optional set of specifications that details how the connection + will be processed after it has been upgraded.¶

+

+ Registrations happen on a "First Come First Served" basis (see + Section 4.4 of [RFC8126]) and are subject to the + following rules:¶

+
    +
  1. A protocol-name token, once registered, stays registered forever.¶ +
    2. +
  2. A protocol-name token is case-insensitive and registered with the + preferred case to be generated by senders.¶ +
    3. +
  3. The registration MUST name a responsible party for the + registration.¶ +
    4. +
  4. The registration MUST name a point of contact.¶ +
    5. +
  5. The registration MAY name a set of specifications associated with + that token. Such specifications need not be publicly available.¶ +
    6. +
  6. The registration SHOULD name a set of expected "protocol-version" + tokens associated with that token at the time of registration.¶ +
    7. +
  7. The responsible party MAY change the registration at any time. + The IANA will keep a record of all such changes, and make them + available upon request.¶ +
    8. +
  8. The IESG MAY reassign responsibility for a protocol token. + This will normally only be used in the case when a + responsible party cannot be contacted.¶ +
    9. +
+
+
+
+
+
+
+

+17. Security Considerations +

+

+ This section is meant to inform developers, information providers, and + users of known security concerns relevant to HTTP semantics and its + use for transferring information over the Internet. Considerations related + to caching are discussed in Section 7 of [CACHING], + and considerations related to HTTP/1.1 message syntax and parsing are + discussed in Section 11 of [HTTP/1.1].¶

+

+ The list of considerations below is not exhaustive. Most security concerns + related to HTTP semantics are about securing server-side applications (code + behind the HTTP interface), securing user agent processing of content + received via HTTP, or secure use of the Internet in general, rather than + security of the protocol. The security considerations for URIs, which + are fundamental to HTTP operation, are discussed in + Section 7 of [URI]. Various organizations maintain + topical information and links to current research on Web application + security (e.g., [OWASP]).¶

+
+
+

+17.1. Establishing Authority +

+ + +

+ HTTP relies on the notion of an "authoritative response": a + response that has been determined by (or at the direction of) the origin + server identified within the target URI to be the most appropriate response + for that request given the state of the target resource at the time of + response message origination.¶

+

+ When a registered name is used in the authority component, the "http" URI + scheme (Section 4.2.1) relies on the user's local name + resolution service to determine where it can find authoritative responses. + This means that any attack on a user's network host table, cached names, + or name resolution libraries becomes an avenue for attack on establishing + authority for "http" URIs. Likewise, the user's choice of server for + Domain Name Service (DNS), and the hierarchy of servers from which it + obtains resolution results, could impact the authenticity of address + mappings; DNS Security Extensions (DNSSEC, [RFC4033]) are + one way to improve authenticity, as are the various mechanisms for making + DNS requests over more secure transfer protocols.¶

+

+ Furthermore, after an IP address is obtained, establishing authority for + an "http" URI is vulnerable to attacks on Internet Protocol routing.¶

+

+ The "https" scheme (Section 4.2.2) is intended to prevent + (or at least reveal) many of these potential attacks on establishing + authority, provided that the negotiated connection is secured and + the client properly verifies that the communicating server's identity + matches the target URI's authority component + (Section 4.3.4). Correctly implementing such verification + can be difficult (see [Georgiev]).¶

+

+ Authority for a given origin server can be delegated through protocol + extensions; for example, [ALTSVC]. Likewise, the set of + servers for which a connection is considered authoritative can be changed + with a protocol extension like [RFC8336].¶

+

+ Providing a response from a non-authoritative source, such as a shared + proxy cache, is often useful to improve performance and availability, but + only to the extent that the source can be trusted or the distrusted + response can be safely used.¶

+

+ Unfortunately, communicating authority to users can be difficult. + For example, "phishing" is an attack on the user's perception + of authority, where that perception can be misled by presenting similar + branding in hypertext, possibly aided by userinfo obfuscating the authority + component (see Section 4.2.1). + User agents can reduce the impact of phishing attacks by enabling users to + easily inspect a target URI prior to making an action, by prominently + distinguishing (or rejecting) userinfo when present, and by not sending + stored credentials and cookies when the referring document is from an + unknown or untrusted source.¶

+
+
+
+
+

+17.2. Risks of Intermediaries +

+

+ HTTP intermediaries are inherently situated for on-path attacks. + Compromise of + the systems on which the intermediaries run can result in serious security + and privacy problems. Intermediaries might have access to security-related + information, personal information about individual users and + organizations, and proprietary information belonging to users and + content providers. A compromised intermediary, or an intermediary + implemented or configured without regard to security and privacy + considerations, might be used in the commission of a wide range of + potential attacks.¶

+

+ Intermediaries that contain a shared cache are especially vulnerable + to cache poisoning attacks, as described in Section 7 of [CACHING].¶

+

+ Implementers need to consider the privacy and security + implications of their design and coding decisions, and of the + configuration options they provide to operators (especially the + default configuration).¶

+

+ Intermediaries are no more trustworthy than the people and policies + under which they operate; HTTP cannot solve this problem.¶

+
+
+
+
+

+17.3. Attacks Based on File and Path Names +

+

+ Origin servers frequently make use of their local file system to manage the + mapping from target URI to resource representations. + Most file systems are not designed to protect against malicious file + or path names. Therefore, an origin server needs to avoid accessing + names that have a special significance to the system when mapping the + target resource to files, folders, or directories.¶

+

+ For example, UNIX, Microsoft Windows, and other operating systems use ".." + as a path component to indicate a directory level above the current one, + and they use specially named paths or file names to send data to system devices. + Similar naming conventions might exist within other types of storage + systems. Likewise, local storage systems have an annoying tendency to + prefer user-friendliness over security when handling invalid or unexpected + characters, recomposition of decomposed characters, and case-normalization + of case-insensitive names.¶

+

+ Attacks based on such special names tend to focus on either denial-of-service + (e.g., telling the server to read from a COM port) or disclosure + of configuration and source files that are not meant to be served.¶

+
+
+
+
+

+17.4. Attacks Based on Command, Code, or Query Injection +

+

+ Origin servers often use parameters within the URI as a + means of identifying system services, selecting database entries, or + choosing a data source. However, data received in a request cannot be + trusted. An attacker could construct any of the request data elements + (method, target URI, header fields, or content) to contain data that might + be misinterpreted as a command, code, or query when passed through a + command invocation, language interpreter, or database interface.¶

+

+ For example, SQL injection is a common attack wherein additional query + language is inserted within some part of the target URI or header + fields (e.g., Host, Referer, etc.). + If the received data is used directly within a SELECT statement, the + query language might be interpreted as a database command instead of a + simple string value. This type of implementation vulnerability is extremely + common, in spite of being easy to prevent.¶

+

+ In general, resource implementations ought to avoid use of request data + in contexts that are processed or interpreted as instructions. Parameters + ought to be compared to fixed strings and acted upon as a result of that + comparison, rather than passed through an interface that is not prepared + for untrusted data. Received data that isn't based on fixed parameters + ought to be carefully filtered or encoded to avoid being misinterpreted.¶

+

+ Similar considerations apply to request data when it is stored and later + processed, such as within log files, monitoring tools, or when included + within a data format that allows embedded scripts.¶

+
+
+
+
+

+17.5. Attacks via Protocol Element Length +

+

+ Because HTTP uses mostly textual, character-delimited fields, parsers are + often vulnerable to attacks based on sending very long (or very slow) + streams of data, particularly where an implementation is expecting a + protocol element with no predefined length + (Section 2.3).¶

+

+ To promote interoperability, specific recommendations are made for minimum + size limits on fields (Section 5.4). These are + minimum recommendations, chosen to be supportable even by implementations + with limited resources; it is expected that most implementations will + choose substantially higher limits.¶

+

+ A server can reject a message that + has a target URI that is too long (Section 15.5.15) or request content + that is too large (Section 15.5.14). Additional status codes related to + capacity limits have been defined by extensions to HTTP + [RFC6585].¶

+

+ Recipients ought to carefully limit the extent to which they process other + protocol elements, including (but not limited to) request methods, response + status phrases, field names, numeric values, and chunk lengths. + Failure to limit such processing can result in arbitrary code execution due to + buffer or arithmetic + overflows, and increased vulnerability to denial-of-service attacks.¶

+
+
+
+
+

+17.6. Attacks Using Shared-Dictionary Compression +

+

+ Some attacks on encrypted protocols use the differences in size created by + dynamic compression to reveal confidential information; for example, [BREACH]. These attacks rely on creating a redundancy between + attacker-controlled content and the confidential information, such that a + dynamic compression algorithm using the same dictionary for both content + will compress more efficiently when the attacker-controlled content matches + parts of the confidential content.¶

+

+ HTTP messages can be compressed in a number of ways, including using TLS + compression, content codings, transfer codings, and other extension or + version-specific mechanisms.¶

+

+ The most effective mitigation for this risk is to disable compression on + sensitive data, or to strictly separate sensitive data from attacker-controlled + data so that they cannot share the same compression dictionary. With + careful design, a compression scheme can be designed in a way that is not + considered exploitable in limited use cases, such as HPACK ([HPACK]).¶

+
+
+
+
+

+17.7. Disclosure of Personal Information +

+

+ Clients are often privy to large amounts of personal information, + including both information provided by the user to interact with resources + (e.g., the user's name, location, mail address, passwords, encryption + keys, etc.) and information about the user's browsing activity over + time (e.g., history, bookmarks, etc.). Implementations need to + prevent unintentional disclosure of personal information.¶

+
+
+
+
+

+17.8. Privacy of Server Log Information +

+

+ A server is in the position to save personal data about a user's requests + over time, which might identify their reading patterns or subjects of + interest. In particular, log information gathered at an intermediary + often contains a history of user agent interaction, across a multitude + of sites, that can be traced to individual users.¶

+

+ HTTP log information is confidential in nature; its handling is often + constrained by laws and regulations. Log information needs to be securely + stored and appropriate guidelines followed for its analysis. + Anonymization of personal information within individual entries helps, + but it is generally not sufficient to prevent real log traces from being + re-identified based on correlation with other access characteristics. + As such, access traces that are keyed to a specific client are unsafe to + publish even if the key is pseudonymous.¶

+

+ To minimize the risk of theft or accidental publication, log information + ought to be purged of personally identifiable information, including + user identifiers, IP addresses, and user-provided query parameters, + as soon as that information is no longer necessary to support operational + needs for security, auditing, or fraud control.¶

+
+
+
+
+

+17.9. Disclosure of Sensitive Information in URIs +

+

+ URIs are intended to be shared, not secured, even when they identify secure + resources. URIs are often shown on displays, added to templates when a page + is printed, and stored in a variety of unprotected bookmark lists. + Many servers, proxies, and user agents log or display the target URI + in places where it might be visible to third parties. + It is therefore unwise to include information within a URI that + is sensitive, personally identifiable, or a risk to disclose.¶

+

+ When an application uses client-side mechanisms to construct a target URI + out of user-provided information, such as the query fields of a form using + GET, potentially sensitive data might be provided that would not be + appropriate for disclosure within a URI. POST is often preferred in such + cases because it usually doesn't construct a URI; instead, POST of a form + transmits the potentially sensitive data in the request content. However, this + hinders caching and uses an unsafe method for what would otherwise be a safe + request. Alternative workarounds include transforming the user-provided data + prior to constructing the URI or filtering the data to only include common + values that are not sensitive. Likewise, redirecting the result of a query + to a different (server-generated) URI can remove potentially sensitive data + from later links and provide a cacheable response for later reuse.¶

+

+ Since the Referer header field tells a target site about the + context that resulted in a request, it has the potential to reveal + information about the user's immediate browsing history and any personal + information that might be found in the referring resource's URI. + Limitations on the Referer header field are described in Section 10.1.3 to + address some of its security considerations.¶

+
+
+
+
+

+17.10. Application Handling of Field Names +

+

+ Servers often use non-HTTP gateway interfaces and frameworks to process a received + request and produce content for the response. For historical reasons, such interfaces + often pass received field names as external variable names, using a name mapping + suitable for environment variables.¶

+

+ For example, the Common Gateway Interface (CGI) mapping of protocol-specific + meta-variables, defined by Section 4.1.18 of [RFC3875], + is applied to received header fields that do not correspond to one of CGI's + standard variables; the mapping consists of prepending "HTTP_" to each name + and changing all instances of hyphen ("-") to underscore ("_"). This same mapping + has been inherited by many other application frameworks in order to simplify + moving applications from one platform to the next.¶

+

+ In CGI, a received Content-Length field would be passed + as the meta-variable "CONTENT_LENGTH" with a string value matching the + received field's value. In contrast, a received "Content_Length" header field would + be passed as the protocol-specific meta-variable "HTTP_CONTENT_LENGTH", + which might lead to some confusion if an application mistakenly reads the + protocol-specific meta-variable instead of the default one. (This historical practice + is why Section 16.3.2.1 discourages the creation + of new field names that contain an underscore.)¶

+

+ Unfortunately, mapping field names to different interface names can lead to + security vulnerabilities if the mapping is incomplete or ambiguous. For example, + if an attacker were to send a field named "Transfer_Encoding", a naive interface + might map that to the same variable name as the "Transfer-Encoding" field, resulting + in a potential request smuggling vulnerability (Section 11.2 of [HTTP/1.1]).¶

+

+ To mitigate the associated risks, implementations that perform such + mappings are advised to make the mapping unambiguous and complete + for the full range of potential octets received as a name (including those + that are discouraged or forbidden by the HTTP grammar). + For example, a field with an unusual name character might + result in the request being blocked, the specific field being removed, + or the name being passed with a different prefix to distinguish it from + other fields.¶

+
+
+
+
+

+17.11. Disclosure of Fragment after Redirects +

+

+ Although fragment identifiers used within URI references are not sent + in requests, implementers ought to be aware that they will be visible to + the user agent and any extensions or scripts running as a result of the + response. In particular, when a redirect occurs and the original request's + fragment identifier is inherited by the new reference in + Location (Section 10.2.2), this might + have the effect of disclosing one site's fragment to another site. + If the first site uses personal information in fragments, it ought to + ensure that redirects to other sites include a (possibly empty) fragment + component in order to block that inheritance.¶

+
+
+
+
+

+17.12. Disclosure of Product Information +

+

+ The User-Agent (Section 10.1.5), + Via (Section 7.6.3), and + Server (Section 10.2.4) header fields often + reveal information about the respective sender's software systems. + In theory, this can make it easier for an attacker to exploit known + security holes; in practice, attackers tend to try all potential holes + regardless of the apparent software versions being used.¶

+

+ Proxies that serve as a portal through a network firewall ought to take + special precautions regarding the transfer of header information that might + identify hosts behind the firewall. The Via header field + allows intermediaries to replace sensitive machine names with pseudonyms.¶

+
+
+
+
+

+17.13. Browser Fingerprinting +

+

+ Browser fingerprinting is a set of techniques for identifying a specific + user agent over time through its unique set of characteristics. These + characteristics might include information related to how it uses the underlying + transport protocol, + feature capabilities, and scripting environment, though of particular + interest here is the set of unique characteristics that might be + communicated via HTTP. Fingerprinting is considered a privacy concern + because it enables tracking of a user agent's behavior over time + ([Bujlow]) without + the corresponding controls that the user might have over other forms of + data collection (e.g., cookies). Many general-purpose user agents + (i.e., Web browsers) have taken steps to reduce their fingerprints.¶

+

+ There are a number of request header fields that might reveal information + to servers that is sufficiently unique to enable fingerprinting. + The From header field is the most obvious, though it is + expected that From will only be sent when self-identification is desired by + the user. Likewise, Cookie header fields are deliberately designed to + enable re-identification, so fingerprinting concerns only apply to + situations where cookies are disabled or restricted by the user agent's + configuration.¶

+

+ The User-Agent header field might contain enough information + to uniquely identify a specific device, usually when combined with other + characteristics, particularly if the user agent sends excessive details + about the user's system or extensions. However, the source of unique + information that is least expected by users is + proactive negotiation (Section 12.1), + including the Accept, Accept-Charset, + Accept-Encoding, and Accept-Language + header fields.¶

+

+ In addition to the fingerprinting concern, detailed use of the + Accept-Language header field can reveal information the + user might consider to be of a private nature. For example, understanding + a given language set might be strongly correlated to membership in a + particular ethnic group. + An approach that limits such loss of privacy would be for a user agent + to omit the sending of Accept-Language except for sites that have been + explicitly permitted, perhaps via interaction after detecting a Vary + header field that indicates language negotiation might be useful.¶

+

+ In environments where proxies are used to enhance privacy, user agents + ought to be conservative in sending proactive negotiation header fields. + General-purpose user agents that provide a high degree of header field + configurability ought to inform users about the loss of privacy that might + result if too much detail is provided. As an extreme privacy measure, + proxies could filter the proactive negotiation header fields in relayed + requests.¶

+
+
+
+
+

+17.14. Validator Retention +

+

+ The validators defined by this specification are not intended to ensure + the validity of a representation, guard against malicious changes, or + detect on-path attacks. At best, they enable more efficient cache + updates and optimistic concurrent writes when all participants are behaving + nicely. At worst, the conditions will fail and the client will receive a + response that is no more harmful than an HTTP exchange without conditional + requests.¶

+

+ An entity tag can be abused in ways that create privacy risks. For example, + a site might deliberately construct a semantically invalid entity tag that + is unique to the user or user agent, send it in a cacheable response with a + long freshness time, and then read that entity tag in later conditional + requests as a means of re-identifying that user or user agent. Such an + identifying tag would become a persistent identifier for as long as the + user agent retained the original cache entry. User agents that cache + representations ought to ensure that the cache is cleared or replaced + whenever the user performs privacy-maintaining actions, such as clearing + stored cookies or changing to a private browsing mode.¶

+
+
+
+
+

+17.15. Denial-of-Service Attacks Using Range +

+

+ Unconstrained multiple range requests are susceptible to denial-of-service + attacks because the effort required to request many overlapping ranges of + the same data is tiny compared to the time, memory, and bandwidth consumed + by attempting to serve the requested data in many parts. + Servers ought to ignore, coalesce, or reject egregious range requests, such + as requests for more than two overlapping ranges or for many small ranges + in a single set, particularly when the ranges are requested out of order + for no apparent reason. Multipart range requests are not designed to + support random access.¶

+
+
+
+
+

+17.16. Authentication Considerations +

+

+ Everything about the topic of HTTP authentication is a security + consideration, so the list of considerations below is not exhaustive. + Furthermore, it is limited to security considerations regarding the + authentication framework, in general, rather than discussing all of the + potential considerations for specific authentication schemes (which ought + to be documented in the specifications that define those schemes). + Various organizations maintain topical information and links to current + research on Web application security (e.g., [OWASP]), + including common pitfalls for implementing and using the authentication + schemes found in practice.¶

+
+
+

+17.16.1. Confidentiality of Credentials +

+

+ The HTTP authentication framework does not define a single mechanism for + maintaining the confidentiality of credentials; instead, each + authentication scheme defines how the credentials are encoded prior to + transmission. While this provides flexibility for the development of future + authentication schemes, it is inadequate for the protection of existing + schemes that provide no confidentiality on their own, or that do not + sufficiently protect against replay attacks. Furthermore, if the server + expects credentials that are specific to each individual user, the exchange + of those credentials will have the effect of identifying that user even if + the content within credentials remains confidential.¶

+

+ HTTP depends on the security properties of the underlying transport- or + session-level connection to provide confidential transmission of + fields. Services that depend on individual user authentication require a + secured connection prior to exchanging credentials + (Section 4.2.2).¶

+
+
+
+
+

+17.16.2. Credentials and Idle Clients +

+

+ Existing HTTP clients and user agents typically retain authentication + information indefinitely. HTTP does not provide a mechanism for the + origin server to direct clients to discard these cached credentials, since + the protocol has no awareness of how credentials are obtained or managed + by the user agent. The mechanisms for expiring or revoking credentials can + be specified as part of an authentication scheme definition.¶

+

+ Circumstances under which credential caching can interfere with the + application's security model include but are not limited to:¶

+
    +
  • Clients that have been idle for an extended period, following + which the server might wish to cause the client to re-prompt the + user for credentials.¶ +
    • +
  • Applications that include a session termination indication + (such as a "logout" or "commit" button on a page) after which + the server side of the application "knows" that there is no + further reason for the client to retain the credentials.¶ +
    • +
+

+ User agents that cache credentials are encouraged to provide a readily + accessible mechanism for discarding cached credentials under user control.¶

+
+
+
+
+

+17.16.3. Protection Spaces +

+

+ Authentication schemes that solely rely on the "realm" mechanism for + establishing a protection space will expose credentials to all resources on + an origin server. Clients that have successfully made authenticated requests + with a resource can use the same authentication credentials for other + resources on the same origin server. This makes it possible for a different + resource to harvest authentication credentials for other resources.¶

+

+ This is of particular concern when an origin server hosts resources for multiple + parties under the same origin (Section 11.5). + Possible mitigation strategies include restricting direct access to + authentication credentials (i.e., not making the content of the + Authorization request header field available), and separating protection + spaces by using a different host name (or port number) for each party.¶

+
+
+
+
+

+17.16.4. Additional Response Fields +

+

+ Adding information to responses that are sent over an unencrypted + channel can affect security and privacy. The presence of the + Authentication-Info and Proxy-Authentication-Info + header fields alone indicates that HTTP authentication is in use. Additional + information could be exposed by the contents of the authentication-scheme + specific parameters; this will have to be considered in the definitions of these + schemes.¶

+
+
+
+
+
+
+
+
+

+18. IANA Considerations +

+

+ The change controller for the following registrations is: + "IETF (iesg@ietf.org) - Internet Engineering Task Force".¶

+
+
+

+18.1. URI Scheme Registration +

+

+ IANA has updated the "Uniform Resource Identifier (URI) Schemes" registry [BCP35] at + <https://www.iana.org/assignments/uri-schemes/> with the + permanent schemes listed in Table 2 in Section 4.2.¶

+
+
+
+
+

+18.2. Method Registration +

+

+ IANA has updated the "Hypertext Transfer Protocol (HTTP) Method Registry" at + <https://www.iana.org/assignments/http-methods> with the + registration procedure of Section 16.1.1 and the method + names summarized in the following table.¶

+
+ + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +
Table 7
MethodSafeIdempotentSection
CONNECTnono + 9.3.6 +
DELETEnoyes + 9.3.5 +
GETyesyes + 9.3.1 +
HEADyesyes + 9.3.2 +
OPTIONSyesyes + 9.3.7 +
POSTnono + 9.3.3 +
PUTnoyes + 9.3.4 +
TRACEyesyes + 9.3.8 +
*nono + 18.2 +
+
+

+ + The method name "*" is reserved because using "*" as a method name would + conflict with its usage as a wildcard in some fields (e.g., + "Access-Control-Request-Method").¶

+
+
+
+
+

+18.3. Status Code Registration +

+

+ IANA has updated the "Hypertext Transfer Protocol (HTTP) Status Code Registry" + at <https://www.iana.org/assignments/http-status-codes> with + the registration procedure of Section 16.2.1 and the + status code values summarized in the following table.¶

+
+ + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +
Table 8
ValueDescriptionSection
100Continue + 15.2.1 +
101Switching Protocols + 15.2.2 +
200OK + 15.3.1 +
201Created + 15.3.2 +
202Accepted + 15.3.3 +
203Non-Authoritative Information + 15.3.4 +
204No Content + 15.3.5 +
205Reset Content + 15.3.6 +
206Partial Content + 15.3.7 +
300Multiple Choices + 15.4.1 +
301Moved Permanently + 15.4.2 +
302Found + 15.4.3 +
303See Other + 15.4.4 +
304Not Modified + 15.4.5 +
305Use Proxy + 15.4.6 +
306(Unused) + 15.4.7 +
307Temporary Redirect + 15.4.8 +
308Permanent Redirect + 15.4.9 +
400Bad Request + 15.5.1 +
401Unauthorized + 15.5.2 +
402Payment Required + 15.5.3 +
403Forbidden + 15.5.4 +
404Not Found + 15.5.5 +
405Method Not Allowed + 15.5.6 +
406Not Acceptable + 15.5.7 +
407Proxy Authentication Required + 15.5.8 +
408Request Timeout + 15.5.9 +
409Conflict + 15.5.10 +
410Gone + 15.5.11 +
411Length Required + 15.5.12 +
412Precondition Failed + 15.5.13 +
413Content Too Large + 15.5.14 +
414URI Too Long + 15.5.15 +
415Unsupported Media Type + 15.5.16 +
416Range Not Satisfiable + 15.5.17 +
417Expectation Failed + 15.5.18 +
418(Unused) + 15.5.19 +
421Misdirected Request + 15.5.20 +
422Unprocessable Content + 15.5.21 +
426Upgrade Required + 15.5.22 +
500Internal Server Error + 15.6.1 +
501Not Implemented + 15.6.2 +
502Bad Gateway + 15.6.3 +
503Service Unavailable + 15.6.4 +
504Gateway Timeout + 15.6.5 +
505HTTP Version Not Supported + 15.6.6 +
+
+
+
+
+
+

+18.4. Field Name Registration +

+

+ This specification updates the HTTP-related aspects of the existing + registration procedures for message header fields defined in [RFC3864]. + It replaces the old procedures as they relate to HTTP by defining a new + registration procedure and moving HTTP field definitions into a separate + registry.¶

+

+ IANA has created a new registry titled "Hypertext Transfer Protocol (HTTP) + Field Name Registry" as outlined in Section 16.3.1.¶

+

+ IANA has moved all entries in the "Permanent Message Header Field + Names" and "Provisional Message Header Field Names" registries (see + <https://www.iana.org/assignments/message-headers/>) with the + protocol 'http' to this registry and has applied the following changes:¶

+
    +
  1. The 'Applicable Protocol' field has been omitted.¶ +
    2. +
  2. Entries that had a status of 'standard', 'experimental', 'reserved', or + 'informational' have been made to have a status of 'permanent'.¶ +
    3. +
  3. Provisional entries without a status have been made to have a status of + 'provisional'.¶ +
    4. +
  4. Permanent entries without a status (after confirmation that the + registration document did not define one) have been made to have a status of + 'provisional'. The expert(s) can choose to update the entries' status if there is + evidence that another is more appropriate.¶ +
    5. +
+

+ IANA has annotated the "Permanent Message Header Field + Names" and "Provisional Message Header Field Names" registries with the + following note to indicate that HTTP field name registrations have moved:¶

+ +

+ IANA has updated the "Hypertext Transfer Protocol (HTTP) Field Name Registry" + with the field names listed in the following table.¶

+
+ + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +
Table 9
Field NameStatusSectionComments
Acceptpermanent + 12.5.1 +
Accept-Charsetdeprecated + 12.5.2 +
Accept-Encodingpermanent + 12.5.3 +
Accept-Languagepermanent + 12.5.4 +
Accept-Rangespermanent + 14.3 +
Allowpermanent + 10.2.1 +
Authentication-Infopermanent + 11.6.3 +
Authorizationpermanent + 11.6.2 +
Connectionpermanent + 7.6.1 +
Content-Encodingpermanent + 8.4 +
Content-Languagepermanent + 8.5 +
Content-Lengthpermanent + 8.6 +
Content-Locationpermanent + 8.7 +
Content-Rangepermanent + 14.4 +
Content-Typepermanent + 8.3 +
Datepermanent + 6.6.1 +
ETagpermanent + 8.8.3 +
Expectpermanent + 10.1.1 +
Frompermanent + 10.1.2 +
Hostpermanent + 7.2 +
If-Matchpermanent + 13.1.1 +
If-Modified-Sincepermanent + 13.1.3 +
If-None-Matchpermanent + 13.1.2 +
If-Rangepermanent + 13.1.5 +
If-Unmodified-Sincepermanent + 13.1.4 +
Last-Modifiedpermanent + 8.8.2 +
Locationpermanent + 10.2.2 +
Max-Forwardspermanent + 7.6.2 +
Proxy-Authenticatepermanent + 11.7.1 +
Proxy-Authentication-Infopermanent + 11.7.3 +
Proxy-Authorizationpermanent + 11.7.2 +
Rangepermanent + 14.2 +
Refererpermanent + 10.1.3 +
Retry-Afterpermanent + 10.2.3 +
Serverpermanent + 10.2.4 +
TEpermanent + 10.1.4 +
Trailerpermanent + 6.6.2 +
Upgradepermanent + 7.8 +
User-Agentpermanent + 10.1.5 +
Varypermanent + 12.5.5 +
Viapermanent + 7.6.3 +
WWW-Authenticatepermanent + 11.6.1 +
*permanent + 12.5.5 + (reserved)
+
+
+

+ + + The field name "*" is reserved because using that name as + an HTTP header field might conflict with its special semantics in the + Vary header field (Section 12.5.5).¶

+
+

+ + + + + IANA has updated the "Content-MD5" entry in the new registry to have + a status of 'obsoleted' with references to + Section 14.15 of [RFC2616] (for the definition + of the header field) and + Appendix B of [RFC7231] (which removed the field + definition from the updated specification).¶

+
+
+
+
+

+18.5. Authentication Scheme Registration +

+

+ IANA has updated the + "Hypertext Transfer Protocol (HTTP) Authentication Scheme Registry" + at <https://www.iana.org/assignments/http-authschemes> with + the registration procedure of Section 16.4.1. + No authentication schemes are defined in this document.¶

+
+
+
+
+

+18.6. Content Coding Registration +

+

+ IANA has updated the "HTTP Content Coding Registry" at + <https://www.iana.org/assignments/http-parameters/> + with the registration procedure of Section 16.6.1 + and the content coding names summarized in the table below.¶

+
+ + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +
Table 10
NameDescriptionSection
compressUNIX "compress" data format [Welch] + + 8.4.1.1 +
deflate"deflate" compressed data ([RFC1951]) inside + the "zlib" data format ([RFC1950]) + 8.4.1.2 +
gzipGZIP file format [RFC1952] + + 8.4.1.3 +
identityReserved + 12.5.3 +
x-compressDeprecated (alias for compress) + 8.4.1.1 +
x-gzipDeprecated (alias for gzip) + 8.4.1.3 +
+
+
+
+
+
+

+18.7. Range Unit Registration +

+

+ IANA has updated the "HTTP Range Unit Registry" at + <https://www.iana.org/assignments/http-parameters/> + with the registration procedure of Section 16.5.1 + and the range unit names summarized in the table below.¶

+
+ + + + + + + + + + + + + + + + + + + + + +
Table 11
Range Unit NameDescriptionSection
bytesa range of octets + 14.1.2 +
nonereserved as keyword to indicate range requests are not supported + 14.3 +
+
+
+
+
+
+

+18.8. Media Type Registration +

+

+ IANA has updated the "Media Types" registry at + <https://www.iana.org/assignments/media-types> + with the registration information in + Section 14.6 + for the media type "multipart/byteranges".¶

+

+ IANA has updated the registry note about "q" parameters with + a link to Section 12.5.1 of this document.¶

+
+
+
+
+

+18.9. Port Registration +

+

+ IANA has updated the "Service Name and Transport Protocol Port Number + Registry" at <https://www.iana.org/assignments/service-names-port-numbers/> + for the services on ports 80 and 443 that use UDP or TCP to:¶

+
    +
  1. use this document as "Reference", and¶ +
    2. +
  2. when currently unspecified, set "Assignee" to "IESG" and "Contact" to + "IETF_Chair".¶ +
    3. +
+
+
+
+
+

+18.10. Upgrade Token Registration +

+

+ IANA has updated the + "Hypertext Transfer Protocol (HTTP) Upgrade Token Registry" at + <https://www.iana.org/assignments/http-upgrade-tokens> + with the registration procedure described in Section 16.7 + and the upgrade token names summarized in the following table.¶

+ + + + + + + + + + + + + + + + + + +
Table 12
NameDescriptionExpected Version TokensSection
HTTPHypertext Transfer Protocolany DIGIT.DIGIT (e.g., "2.0") + 2.5 +
+
+
+
+
+
+

+19. References +

+
+

+19.1. Normative References +

+
+
[CACHING]
+
+Fielding, R., Ed., Nottingham, M., Ed., and J. Reschke, Ed., "HTTP Caching", STD 98, RFC 9111, DOI 10.17487/RFC9111, , <https://www.rfc-editor.org/info/rfc9111>.
+
+
[RFC1950]
+
+Deutsch, P. and J-L. Gailly, "ZLIB Compressed Data Format Specification version 3.3", RFC 1950, DOI 10.17487/RFC1950, , <https://www.rfc-editor.org/info/rfc1950>.
+
+
[RFC1951]
+
+Deutsch, P., "DEFLATE Compressed Data Format Specification version 1.3", RFC 1951, DOI 10.17487/RFC1951, , <https://www.rfc-editor.org/info/rfc1951>.
+
+
[RFC1952]
+
+Deutsch, P., "GZIP file format specification version 4.3", RFC 1952, DOI 10.17487/RFC1952, , <https://www.rfc-editor.org/info/rfc1952>.
+
+
[RFC2046]
+
+Freed, N. and N. Borenstein, "Multipurpose Internet Mail Extensions (MIME) Part Two: Media Types", RFC 2046, DOI 10.17487/RFC2046, , <https://www.rfc-editor.org/info/rfc2046>.
+
+
[RFC2119]
+
+Bradner, S., "Key words for use in RFCs to Indicate Requirement Levels", BCP 14, RFC 2119, DOI 10.17487/RFC2119, , <https://www.rfc-editor.org/info/rfc2119>.
+
+
[RFC4647]
+
+Phillips, A., Ed. and M. Davis, Ed., "Matching of Language Tags", BCP 47, RFC 4647, DOI 10.17487/RFC4647, , <https://www.rfc-editor.org/info/rfc4647>.
+
+
[RFC4648]
+
+Josefsson, S., "The Base16, Base32, and Base64 Data Encodings", RFC 4648, DOI 10.17487/RFC4648, , <https://www.rfc-editor.org/info/rfc4648>.
+
+
[RFC5234]
+
+Crocker, D., Ed. and P. Overell, "Augmented BNF for Syntax Specifications: ABNF", STD 68, RFC 5234, DOI 10.17487/RFC5234, , <https://www.rfc-editor.org/info/rfc5234>.
+
+
[RFC5280]
+
+Cooper, D., Santesson, S., Farrell, S., Boeyen, S., Housley, R., and W. Polk, "Internet X.509 Public Key Infrastructure Certificate and Certificate Revocation List (CRL) Profile", RFC 5280, DOI 10.17487/RFC5280, , <https://www.rfc-editor.org/info/rfc5280>.
+
+
[RFC5322]
+
+Resnick, P., Ed., "Internet Message Format", RFC 5322, DOI 10.17487/RFC5322, , <https://www.rfc-editor.org/info/rfc5322>.
+
+
[RFC5646]
+
+Phillips, A., Ed. and M. Davis, Ed., "Tags for Identifying Languages", BCP 47, RFC 5646, DOI 10.17487/RFC5646, , <https://www.rfc-editor.org/info/rfc5646>.
+
+
[RFC6125]
+
+Saint-Andre, P. and J. Hodges, "Representation and Verification of Domain-Based Application Service Identity within Internet Public Key Infrastructure Using X.509 (PKIX) Certificates in the Context of Transport Layer Security (TLS)", RFC 6125, DOI 10.17487/RFC6125, , <https://www.rfc-editor.org/info/rfc6125>.
+
+
[RFC6365]
+
+Hoffman, P. and J. Klensin, "Terminology Used in Internationalization in the IETF", BCP 166, RFC 6365, DOI 10.17487/RFC6365, , <https://www.rfc-editor.org/info/rfc6365>.
+
+
[RFC7405]
+
+Kyzivat, P., "Case-Sensitive String Support in ABNF", RFC 7405, DOI 10.17487/RFC7405, , <https://www.rfc-editor.org/info/rfc7405>.
+
+
[RFC8174]
+
+Leiba, B., "Ambiguity of Uppercase vs Lowercase in RFC 2119 Key Words", BCP 14, RFC 8174, DOI 10.17487/RFC8174, , <https://www.rfc-editor.org/info/rfc8174>.
+
+
[TCP]
+
+Postel, J., "Transmission Control Protocol", STD 7, RFC 793, DOI 10.17487/RFC0793, , <https://www.rfc-editor.org/info/rfc793>.
+
+
[TLS13]
+
+Rescorla, E., "The Transport Layer Security (TLS) Protocol Version 1.3", RFC 8446, DOI 10.17487/RFC8446, , <https://www.rfc-editor.org/info/rfc8446>.
+
+
[URI]
+
+Berners-Lee, T., Fielding, R., and L. Masinter, "Uniform Resource Identifier (URI): Generic Syntax", STD 66, RFC 3986, DOI 10.17487/RFC3986, , <https://www.rfc-editor.org/info/rfc3986>.
+
+
[USASCII]
+
+American National Standards Institute, "Coded Character Set -- 7-bit American Standard Code for Information Interchange", ANSI X3.4, .
+
+
[Welch]
+
+Welch, T., "A Technique for High-Performance Data Compression", IEEE Computer 17(6), DOI 10.1109/MC.1984.1659158, , <https://ieeexplore.ieee.org/document/1659158/>.
+
+
+
+
+

+19.2. Informative References +

+
+
[ALTSVC]
+
+Nottingham, M., McManus, P., and J. Reschke, "HTTP Alternative Services", RFC 7838, DOI 10.17487/RFC7838, , <https://www.rfc-editor.org/info/rfc7838>.
+
+
[BCP13]
+
+
+ Freed, N. and J. Klensin, "Multipurpose Internet Mail Extensions (MIME) Part Four: Registration Procedures", BCP 13, RFC 4289, .
+
+ Freed, N., Klensin, J., and T. Hansen, "Media Type Specifications and Registration Procedures", BCP 13, RFC 6838, .
+<https://www.rfc-editor.org/info/bcp13> +
+
+
[BCP178]
+
+
+ Saint-Andre, P., Crocker, D., and M. Nottingham, "Deprecating the "X-" Prefix and Similar Constructs in Application Protocols", BCP 178, RFC 6648, .
+<https://www.rfc-editor.org/info/bcp178> +
+
+
[BCP35]
+
+
+ Thaler, D., Ed., Hansen, T., and T. Hardie, "Guidelines and Registration Procedures for URI Schemes", BCP 35, RFC 7595, .
+<https://www.rfc-editor.org/info/bcp35> +
+
+
[BREACH]
+
+Gluck, Y., Harris, N., and A. Prado, "BREACH: Reviving the CRIME Attack", , <http://breachattack.com/resources/BREACH%20-%20SSL,%20gone%20in%2030%20seconds.pdf>.
+
+
[Bujlow]
+
+Bujlow, T., Carela-Español, V., Solé-Pareta, J., and P. Barlet-Ros, "A Survey on Web Tracking: Mechanisms, Implications, and Defenses", In Proceedings of the IEEE 105(8), DOI 10.1109/JPROC.2016.2637878, , <https://doi.org/10.1109/JPROC.2016.2637878>.
+
+ +
+Barth, A., "HTTP State Management Mechanism", RFC 6265, DOI 10.17487/RFC6265, , <https://www.rfc-editor.org/info/rfc6265>.
+
+
[Err1912]
+
+RFC Errata, Erratum ID 1912, RFC 2978, <https://www.rfc-editor.org/errata/eid1912>.
+
+
[Err5433]
+
+RFC Errata, Erratum ID 5433, RFC 2978, <https://www.rfc-editor.org/errata/eid5433>.
+
+
[Georgiev]
+
+Georgiev, M., Iyengar, S., Jana, S., Anubhai, R., Boneh, D., and V. Shmatikov, "The Most Dangerous Code in the World: Validating SSL Certificates in Non-Browser Software", In Proceedings of the 2012 ACM Conference on Computer and Communications Security (CCS '12), pp. 38-49, DOI 10.1145/2382196.2382204, , <https://doi.org/10.1145/2382196.2382204>.
+
+
[HPACK]
+
+Peon, R. and H. Ruellan, "HPACK: Header Compression for HTTP/2", RFC 7541, DOI 10.17487/RFC7541, , <https://www.rfc-editor.org/info/rfc7541>.
+
+
[HTTP/1.0]
+
+Berners-Lee, T., Fielding, R., and H. Frystyk, "Hypertext Transfer Protocol -- HTTP/1.0", RFC 1945, DOI 10.17487/RFC1945, , <https://www.rfc-editor.org/info/rfc1945>.
+
+
[HTTP/1.1]
+
+Fielding, R., Ed., Nottingham, M., Ed., and J. Reschke, Ed., "HTTP/1.1", STD 99, RFC 9112, DOI 10.17487/RFC9112, , <https://www.rfc-editor.org/info/rfc9112>.
+
+
[HTTP/2]
+
+Thomson, M., Ed. and C. Benfield, Ed., "HTTP/2", RFC 9113, DOI 10.17487/RFC9113, , <https://www.rfc-editor.org/info/rfc9113>.
+
+
[HTTP/3]
+
+Bishop, M., Ed., "HTTP/3", RFC 9114, DOI 10.17487/RFC9114, , <https://www.rfc-editor.org/info/rfc9114>.
+
+
[ISO-8859-1]
+
+International Organization for Standardization, "Information technology -- 8-bit single-byte coded graphic character sets -- Part 1: Latin alphabet No. 1", ISO/IEC 8859-1:1998, .
+
+
[Kri2001]
+
+Kristol, D., "HTTP Cookies: Standards, Privacy, and Politics", ACM Transactions on Internet Technology 1(2), , <http://arxiv.org/abs/cs.SE/0105018>.
+
+
[OWASP]
+
+The Open Web Application Security Project, <https://www.owasp.org/>.
+
+
[REST]
+
+Fielding, R.T., "Architectural Styles and the Design of Network-based Software Architectures", Doctoral Dissertation, University of California, Irvine, , <https://roy.gbiv.com/pubs/dissertation/top.htm>.
+
+
[RFC1919]
+
+Chatel, M., "Classical versus Transparent IP Proxies", RFC 1919, DOI 10.17487/RFC1919, , <https://www.rfc-editor.org/info/rfc1919>.
+
+
[RFC2047]
+
+Moore, K., "MIME (Multipurpose Internet Mail Extensions) Part Three: Message Header Extensions for Non-ASCII Text", RFC 2047, DOI 10.17487/RFC2047, , <https://www.rfc-editor.org/info/rfc2047>.
+
+
[RFC2068]
+
+Fielding, R., Gettys, J., Mogul, J., Frystyk, H., and T. Berners-Lee, "Hypertext Transfer Protocol -- HTTP/1.1", RFC 2068, DOI 10.17487/RFC2068, , <https://www.rfc-editor.org/info/rfc2068>.
+
+
[RFC2145]
+
+Mogul, J. C., Fielding, R., Gettys, J., and H. Frystyk, "Use and Interpretation of HTTP Version Numbers", RFC 2145, DOI 10.17487/RFC2145, , <https://www.rfc-editor.org/info/rfc2145>.
+
+
[RFC2295]
+
+Holtman, K. and A. Mutz, "Transparent Content Negotiation in HTTP", RFC 2295, DOI 10.17487/RFC2295, , <https://www.rfc-editor.org/info/rfc2295>.
+
+
[RFC2324]
+
+Masinter, L., "Hyper Text Coffee Pot Control Protocol (HTCPCP/1.0)", RFC 2324, DOI 10.17487/RFC2324, , <https://www.rfc-editor.org/info/rfc2324>.
+
+
[RFC2557]
+
+Palme, J., Hopmann, A., and N. Shelness, "MIME Encapsulation of Aggregate Documents, such as HTML (MHTML)", RFC 2557, DOI 10.17487/RFC2557, , <https://www.rfc-editor.org/info/rfc2557>.
+
+
[RFC2616]
+
+Fielding, R., Gettys, J., Mogul, J., Frystyk, H., Masinter, L., Leach, P., and T. Berners-Lee, "Hypertext Transfer Protocol -- HTTP/1.1", RFC 2616, DOI 10.17487/RFC2616, , <https://www.rfc-editor.org/info/rfc2616>.
+
+
[RFC2617]
+
+Franks, J., Hallam-Baker, P., Hostetler, J., Lawrence, S., Leach, P., Luotonen, A., and L. Stewart, "HTTP Authentication: Basic and Digest Access Authentication", RFC 2617, DOI 10.17487/RFC2617, , <https://www.rfc-editor.org/info/rfc2617>.
+
+
[RFC2774]
+
+Nielsen, H., Leach, P., and S. Lawrence, "An HTTP Extension Framework", RFC 2774, DOI 10.17487/RFC2774, , <https://www.rfc-editor.org/info/rfc2774>.
+
+
[RFC2818]
+
+Rescorla, E., "HTTP Over TLS", RFC 2818, DOI 10.17487/RFC2818, , <https://www.rfc-editor.org/info/rfc2818>.
+
+
[RFC2978]
+
+Freed, N. and J. Postel, "IANA Charset Registration Procedures", BCP 19, RFC 2978, DOI 10.17487/RFC2978, , <https://www.rfc-editor.org/info/rfc2978>.
+
+
[RFC3040]
+
+Cooper, I., Melve, I., and G. Tomlinson, "Internet Web Replication and Caching Taxonomy", RFC 3040, DOI 10.17487/RFC3040, , <https://www.rfc-editor.org/info/rfc3040>.
+
+
[RFC3864]
+
+Klyne, G., Nottingham, M., and J. Mogul, "Registration Procedures for Message Header Fields", BCP 90, RFC 3864, DOI 10.17487/RFC3864, , <https://www.rfc-editor.org/info/rfc3864>.
+
+
[RFC3875]
+
+Robinson, D. and K. Coar, "The Common Gateway Interface (CGI) Version 1.1", RFC 3875, DOI 10.17487/RFC3875, , <https://www.rfc-editor.org/info/rfc3875>.
+
+
[RFC4033]
+
+Arends, R., Austein, R., Larson, M., Massey, D., and S. Rose, "DNS Security Introduction and Requirements", RFC 4033, DOI 10.17487/RFC4033, , <https://www.rfc-editor.org/info/rfc4033>.
+
+
[RFC4559]
+
+Jaganathan, K., Zhu, L., and J. Brezak, "SPNEGO-based Kerberos and NTLM HTTP Authentication in Microsoft Windows", RFC 4559, DOI 10.17487/RFC4559, , <https://www.rfc-editor.org/info/rfc4559>.
+
+
[RFC5789]
+
+Dusseault, L. and J. Snell, "PATCH Method for HTTP", RFC 5789, DOI 10.17487/RFC5789, , <https://www.rfc-editor.org/info/rfc5789>.
+
+
[RFC5905]
+
+Mills, D., Martin, J., Ed., Burbank, J., and W. Kasch, "Network Time Protocol Version 4: Protocol and Algorithms Specification", RFC 5905, DOI 10.17487/RFC5905, , <https://www.rfc-editor.org/info/rfc5905>.
+
+
[RFC6454]
+
+Barth, A., "The Web Origin Concept", RFC 6454, DOI 10.17487/RFC6454, , <https://www.rfc-editor.org/info/rfc6454>.
+
+
[RFC6585]
+
+Nottingham, M. and R. Fielding, "Additional HTTP Status Codes", RFC 6585, DOI 10.17487/RFC6585, , <https://www.rfc-editor.org/info/rfc6585>.
+
+
[RFC7230]
+
+Fielding, R., Ed. and J. Reschke, Ed., "Hypertext Transfer Protocol (HTTP/1.1): Message Syntax and Routing", RFC 7230, DOI 10.17487/RFC7230, , <https://www.rfc-editor.org/info/rfc7230>.
+
+
[RFC7231]
+
+Fielding, R., Ed. and J. Reschke, Ed., "Hypertext Transfer Protocol (HTTP/1.1): Semantics and Content", RFC 7231, DOI 10.17487/RFC7231, , <https://www.rfc-editor.org/info/rfc7231>.
+
+
[RFC7232]
+
+Fielding, R., Ed. and J. Reschke, Ed., "Hypertext Transfer Protocol (HTTP/1.1): Conditional Requests", RFC 7232, DOI 10.17487/RFC7232, , <https://www.rfc-editor.org/info/rfc7232>.
+
+
[RFC7233]
+
+Fielding, R., Ed., Lafon, Y., Ed., and J. Reschke, Ed., "Hypertext Transfer Protocol (HTTP/1.1): Range Requests", RFC 7233, DOI 10.17487/RFC7233, , <https://www.rfc-editor.org/info/rfc7233>.
+
+
[RFC7234]
+
+Fielding, R., Ed., Nottingham, M., Ed., and J. Reschke, Ed., "Hypertext Transfer Protocol (HTTP/1.1): Caching", RFC 7234, DOI 10.17487/RFC7234, , <https://www.rfc-editor.org/info/rfc7234>.
+
+
[RFC7235]
+
+Fielding, R., Ed. and J. Reschke, Ed., "Hypertext Transfer Protocol (HTTP/1.1): Authentication", RFC 7235, DOI 10.17487/RFC7235, , <https://www.rfc-editor.org/info/rfc7235>.
+
+
[RFC7538]
+
+Reschke, J., "The Hypertext Transfer Protocol Status Code 308 (Permanent Redirect)", RFC 7538, DOI 10.17487/RFC7538, , <https://www.rfc-editor.org/info/rfc7538>.
+
+
[RFC7540]
+
+Belshe, M., Peon, R., and M. Thomson, Ed., "Hypertext Transfer Protocol Version 2 (HTTP/2)", RFC 7540, DOI 10.17487/RFC7540, , <https://www.rfc-editor.org/info/rfc7540>.
+
+
[RFC7578]
+
+Masinter, L., "Returning Values from Forms: multipart/form-data", RFC 7578, DOI 10.17487/RFC7578, , <https://www.rfc-editor.org/info/rfc7578>.
+
+
[RFC7615]
+
+Reschke, J., "HTTP Authentication-Info and Proxy-Authentication-Info Response Header Fields", RFC 7615, DOI 10.17487/RFC7615, , <https://www.rfc-editor.org/info/rfc7615>.
+
+
[RFC7616]
+
+Shekh-Yusef, R., Ed., Ahrens, D., and S. Bremer, "HTTP Digest Access Authentication", RFC 7616, DOI 10.17487/RFC7616, , <https://www.rfc-editor.org/info/rfc7616>.
+
+
[RFC7617]
+
+Reschke, J., "The 'Basic' HTTP Authentication Scheme", RFC 7617, DOI 10.17487/RFC7617, , <https://www.rfc-editor.org/info/rfc7617>.
+
+
[RFC7694]
+
+Reschke, J., "Hypertext Transfer Protocol (HTTP) Client-Initiated Content-Encoding", RFC 7694, DOI 10.17487/RFC7694, , <https://www.rfc-editor.org/info/rfc7694>.
+
+
[RFC8126]
+
+Cotton, M., Leiba, B., and T. Narten, "Guidelines for Writing an IANA Considerations Section in RFCs", BCP 26, RFC 8126, DOI 10.17487/RFC8126, , <https://www.rfc-editor.org/info/rfc8126>.
+
+
[RFC8187]
+
+Reschke, J., "Indicating Character Encoding and Language for HTTP Header Field Parameters", RFC 8187, DOI 10.17487/RFC8187, , <https://www.rfc-editor.org/info/rfc8187>.
+
+
[RFC8246]
+
+McManus, P., "HTTP Immutable Responses", RFC 8246, DOI 10.17487/RFC8246, , <https://www.rfc-editor.org/info/rfc8246>.
+
+
[RFC8288]
+
+Nottingham, M., "Web Linking", RFC 8288, DOI 10.17487/RFC8288, , <https://www.rfc-editor.org/info/rfc8288>.
+
+
[RFC8336]
+
+Nottingham, M. and E. Nygren, "The ORIGIN HTTP/2 Frame", RFC 8336, DOI 10.17487/RFC8336, , <https://www.rfc-editor.org/info/rfc8336>.
+
+
[RFC8615]
+
+Nottingham, M., "Well-Known Uniform Resource Identifiers (URIs)", RFC 8615, DOI 10.17487/RFC8615, , <https://www.rfc-editor.org/info/rfc8615>.
+
+
[RFC8941]
+
+Nottingham, M. and P-H. Kamp, "Structured Field Values for HTTP", RFC 8941, DOI 10.17487/RFC8941, , <https://www.rfc-editor.org/info/rfc8941>.
+
+
[Sniffing]
+
+WHATWG, "MIME Sniffing", <https://mimesniff.spec.whatwg.org>.
+
+
[WEBDAV]
+
+Dusseault, L., Ed., "HTTP Extensions for Web Distributed Authoring and Versioning (WebDAV)", RFC 4918, DOI 10.17487/RFC4918, , <https://www.rfc-editor.org/info/rfc4918>.
+
+
+
+
+
+
+

+Appendix A. Collected ABNF +

+

In the collected ABNF below, list rules are expanded per Section 5.6.1.¶

+
+
Accept = [ ( media-range [ weight ] ) *( OWS "," OWS ( media-range [
+ weight ] ) ) ]
+Accept-Charset = [ ( ( token / "*" ) [ weight ] ) *( OWS "," OWS ( (
+ token / "*" ) [ weight ] ) ) ]
+Accept-Encoding = [ ( codings [ weight ] ) *( OWS "," OWS ( codings [
+ weight ] ) ) ]
+Accept-Language = [ ( language-range [ weight ] ) *( OWS "," OWS (
+ language-range [ weight ] ) ) ]
+Accept-Ranges = acceptable-ranges
+Allow = [ method *( OWS "," OWS method ) ]
+Authentication-Info = [ auth-param *( OWS "," OWS auth-param ) ]
+Authorization = credentials
+
+BWS = OWS
+
+Connection = [ connection-option *( OWS "," OWS connection-option )
+ ]
+Content-Encoding = [ content-coding *( OWS "," OWS content-coding )
+ ]
+Content-Language = [ language-tag *( OWS "," OWS language-tag ) ]
+Content-Length = 1*DIGIT
+Content-Location = absolute-URI / partial-URI
+Content-Range = range-unit SP ( range-resp / unsatisfied-range )
+Content-Type = media-type
+
+Date = HTTP-date
+
+ETag = entity-tag
+Expect = [ expectation *( OWS "," OWS expectation ) ]
+
+From = mailbox
+
+GMT = %x47.4D.54 ; GMT
+
+HTTP-date = IMF-fixdate / obs-date
+Host = uri-host [ ":" port ]
+
+IMF-fixdate = day-name "," SP date1 SP time-of-day SP GMT
+If-Match = "*" / [ entity-tag *( OWS "," OWS entity-tag ) ]
+If-Modified-Since = HTTP-date
+If-None-Match = "*" / [ entity-tag *( OWS "," OWS entity-tag ) ]
+If-Range = entity-tag / HTTP-date
+If-Unmodified-Since = HTTP-date
+
+Last-Modified = HTTP-date
+Location = URI-reference
+
+Max-Forwards = 1*DIGIT
+
+OWS = *( SP / HTAB )
+
+Proxy-Authenticate = [ challenge *( OWS "," OWS challenge ) ]
+Proxy-Authentication-Info = [ auth-param *( OWS "," OWS auth-param )
+ ]
+Proxy-Authorization = credentials
+
+RWS = 1*( SP / HTAB )
+Range = ranges-specifier
+Referer = absolute-URI / partial-URI
+Retry-After = HTTP-date / delay-seconds
+
+Server = product *( RWS ( product / comment ) )
+
+TE = [ t-codings *( OWS "," OWS t-codings ) ]
+Trailer = [ field-name *( OWS "," OWS field-name ) ]
+
+URI-reference = <URI-reference, see [URI], Section 4.1>
+Upgrade = [ protocol *( OWS "," OWS protocol ) ]
+User-Agent = product *( RWS ( product / comment ) )
+
+Vary = [ ( "*" / field-name ) *( OWS "," OWS ( "*" / field-name ) )
+ ]
+Via = [ ( received-protocol RWS received-by [ RWS comment ] ) *( OWS
+ "," OWS ( received-protocol RWS received-by [ RWS comment ] ) ) ]
+
+WWW-Authenticate = [ challenge *( OWS "," OWS challenge ) ]
+
+absolute-URI = <absolute-URI, see [URI], Section 4.3>
+absolute-path = 1*( "/" segment )
+acceptable-ranges = range-unit *( OWS "," OWS range-unit )
+asctime-date = day-name SP date3 SP time-of-day SP year
+auth-param = token BWS "=" BWS ( token / quoted-string )
+auth-scheme = token
+authority = <authority, see [URI], Section 3.2>
+
+challenge = auth-scheme [ 1*SP ( token68 / [ auth-param *( OWS ","
+ OWS auth-param ) ] ) ]
+codings = content-coding / "identity" / "*"
+comment = "(" *( ctext / quoted-pair / comment ) ")"
+complete-length = 1*DIGIT
+connection-option = token
+content-coding = token
+credentials = auth-scheme [ 1*SP ( token68 / [ auth-param *( OWS ","
+ OWS auth-param ) ] ) ]
+ctext = HTAB / SP / %x21-27 ; '!'-'''
+ / %x2A-5B ; '*'-'['
+ / %x5D-7E ; ']'-'~'
+ / obs-text
+
+date1 = day SP month SP year
+date2 = day "-" month "-" 2DIGIT
+date3 = month SP ( 2DIGIT / ( SP DIGIT ) )
+day = 2DIGIT
+day-name = %x4D.6F.6E ; Mon
+ / %x54.75.65 ; Tue
+ / %x57.65.64 ; Wed
+ / %x54.68.75 ; Thu
+ / %x46.72.69 ; Fri
+ / %x53.61.74 ; Sat
+ / %x53.75.6E ; Sun
+day-name-l = %x4D.6F.6E.64.61.79 ; Monday
+ / %x54.75.65.73.64.61.79 ; Tuesday
+ / %x57.65.64.6E.65.73.64.61.79 ; Wednesday
+ / %x54.68.75.72.73.64.61.79 ; Thursday
+ / %x46.72.69.64.61.79 ; Friday
+ / %x53.61.74.75.72.64.61.79 ; Saturday
+ / %x53.75.6E.64.61.79 ; Sunday
+delay-seconds = 1*DIGIT
+
+entity-tag = [ weak ] opaque-tag
+etagc = "!" / %x23-7E ; '#'-'~'
+ / obs-text
+expectation = token [ "=" ( token / quoted-string ) parameters ]
+
+field-content = field-vchar [ 1*( SP / HTAB / field-vchar )
+ field-vchar ]
+field-name = token
+field-value = *field-content
+field-vchar = VCHAR / obs-text
+first-pos = 1*DIGIT
+
+hour = 2DIGIT
+http-URI = "http://" authority path-abempty [ "?" query ]
+https-URI = "https://" authority path-abempty [ "?" query ]
+
+incl-range = first-pos "-" last-pos
+int-range = first-pos "-" [ last-pos ]
+
+language-range = <language-range, see [RFC4647], Section 2.1>
+language-tag = <Language-Tag, see [RFC5646], Section 2.1>
+last-pos = 1*DIGIT
+
+mailbox = <mailbox, see [RFC5322], Section 3.4>
+media-range = ( "*/*" / ( type "/*" ) / ( type "/" subtype ) )
+ parameters
+media-type = type "/" subtype parameters
+method = token
+minute = 2DIGIT
+month = %x4A.61.6E ; Jan
+ / %x46.65.62 ; Feb
+ / %x4D.61.72 ; Mar
+ / %x41.70.72 ; Apr
+ / %x4D.61.79 ; May
+ / %x4A.75.6E ; Jun
+ / %x4A.75.6C ; Jul
+ / %x41.75.67 ; Aug
+ / %x53.65.70 ; Sep
+ / %x4F.63.74 ; Oct
+ / %x4E.6F.76 ; Nov
+ / %x44.65.63 ; Dec
+
+obs-date = rfc850-date / asctime-date
+obs-text = %x80-FF
+opaque-tag = DQUOTE *etagc DQUOTE
+other-range = 1*( %x21-2B ; '!'-'+'
+ / %x2D-7E ; '-'-'~'
+ )
+
+parameter = parameter-name "=" parameter-value
+parameter-name = token
+parameter-value = ( token / quoted-string )
+parameters = *( OWS ";" OWS [ parameter ] )
+partial-URI = relative-part [ "?" query ]
+path-abempty = <path-abempty, see [URI], Section 3.3>
+port = <port, see [URI], Section 3.2.3>
+product = token [ "/" product-version ]
+product-version = token
+protocol = protocol-name [ "/" protocol-version ]
+protocol-name = token
+protocol-version = token
+pseudonym = token
+
+qdtext = HTAB / SP / "!" / %x23-5B ; '#'-'['
+ / %x5D-7E ; ']'-'~'
+ / obs-text
+query = <query, see [URI], Section 3.4>
+quoted-pair = "\" ( HTAB / SP / VCHAR / obs-text )
+quoted-string = DQUOTE *( qdtext / quoted-pair ) DQUOTE
+qvalue = ( "0" [ "." *3DIGIT ] ) / ( "1" [ "." *3"0" ] )
+
+range-resp = incl-range "/" ( complete-length / "*" )
+range-set = range-spec *( OWS "," OWS range-spec )
+range-spec = int-range / suffix-range / other-range
+range-unit = token
+ranges-specifier = range-unit "=" range-set
+received-by = pseudonym [ ":" port ]
+received-protocol = [ protocol-name "/" ] protocol-version
+relative-part = <relative-part, see [URI], Section 4.2>
+rfc850-date = day-name-l "," SP date2 SP time-of-day SP GMT
+
+second = 2DIGIT
+segment = <segment, see [URI], Section 3.3>
+subtype = token
+suffix-length = 1*DIGIT
+suffix-range = "-" suffix-length
+
+t-codings = "trailers" / ( transfer-coding [ weight ] )
+tchar = "!" / "#" / "$" / "%" / "&" / "'" / "*" / "+" / "-" / "." /
+ "^" / "_" / "`" / "|" / "~" / DIGIT / ALPHA
+time-of-day = hour ":" minute ":" second
+token = 1*tchar
+token68 = 1*( ALPHA / DIGIT / "-" / "." / "_" / "~" / "+" / "/" )
+ *"="
+transfer-coding = token *( OWS ";" OWS transfer-parameter )
+transfer-parameter = token BWS "=" BWS ( token / quoted-string )
+type = token
+
+unsatisfied-range = "*/" complete-length
+uri-host = <host, see [URI], Section 3.2.2>
+
+weak = %x57.2F ; W/
+weight = OWS ";" OWS "q=" qvalue
+
+year = 4DIGIT
+
¶ +
+
+
+
+
+

+Appendix B. Changes from Previous RFCs +

+
+
+

+B.1. Changes from RFC 2818 +

+

+ None.¶

+
+
+
+
+

+B.2. Changes from RFC 7230 +

+

+ The sections introducing HTTP's design goals, history, architecture, + conformance criteria, protocol versioning, URIs, message routing, and + header fields have been moved here.¶

+

+ The requirement on semantic conformance has been replaced with permission to + ignore or work around implementation-specific failures. + (Section 2.2)¶

+

+ The description of an origin and authoritative access to origin servers has + been extended for both "http" and "https" URIs to account for alternative + services and secured connections that are not necessarily based on TCP. + (Sections 4.2.1, 4.2.2, + 4.3.1, and 7.3.3)¶

+

+ Explicit requirements have been added to check the target URI scheme's semantics + and reject requests that don't meet any associated requirements. + (Section 7.4)¶

+

+ Parameters in media type, media range, and expectation can be empty via + one or more trailing semicolons. + (Section 5.6.6)¶

+

+ "Field value" now refers to the value after multiple field lines are combined + with commas -- by far the most common use. To refer to a single header + line's value, use "field line value". + (Section 6.3)¶

+

+ Trailer field semantics now transcend the specifics of chunked transfer coding. + The use of trailer fields has been further limited to allow generation + as a trailer field only when the sender knows the field defines that usage and + to allow merging into the header section only if the recipient knows the + corresponding field definition permits and defines how to merge. In all + other cases, implementations are encouraged either to store the trailer + fields separately or to discard them instead of merging. + (Section 6.5.1)¶

+

+ The priority of the absolute form of the request URI over the Host + header field by origin servers has been made explicit to align with proxy handling. + (Section 7.2)¶

+

+ The grammar definition for the Via field's "received-by" was + expanded in RFC 7230 due to changes in the URI grammar for host + [URI] that are not desirable for Via. For simplicity, + we have removed uri-host from the received-by production because it can + be encompassed by the existing grammar for pseudonym. In particular, this + change removed comma from the allowed set of characters for a host name in + received-by. + (Section 7.6.3)¶

+
+
+
+
+

+B.3. Changes from RFC 7231 +

+

+ Minimum URI lengths to be supported by implementations are now recommended. + (Section 4.1)¶

+

+ The following have been clarified: CR and NUL in field values are to be rejected or + mapped to SP, and leading and trailing whitespace needs to be + stripped from field values before they are consumed. + (Section 5.5)¶

+

+ Parameters in media type, media range, and expectation can be empty via + one or more trailing semicolons. + (Section 5.6.6)¶

+

+ An abstract data type for HTTP messages has been introduced to define the + components of a message and their semantics as an abstraction across + multiple HTTP versions, rather than in terms of the specific syntax form of + HTTP/1.1 in [HTTP/1.1], and reflect the contents after the + message is parsed. This makes it easier to distinguish between requirements + on the content (what is conveyed) versus requirements on the messaging + syntax (how it is conveyed) and avoids baking limitations of early protocol + versions into the future of HTTP. (Section 6)¶

+

+ The terms "payload" and "payload body" have been replaced with "content", to better + align with its usage elsewhere (e.g., in field names) and to avoid confusion + with frame payloads in HTTP/2 and HTTP/3. + (Section 6.4)¶

+

+ The term "effective request URI" has been replaced with "target URI". + (Section 7.1)¶

+

+ Restrictions on client retries have been loosened to reflect implementation + behavior. + (Section 9.2.2)¶

+

+ The fact that request bodies on GET, HEAD, and DELETE are not interoperable has been clarified. + (Sections 9.3.1, 9.3.2, and 9.3.5)¶

+

+ The use of the Content-Range header field + (Section 14.4) as a request modifier on PUT is allowed. + (Section 9.3.4)¶

+

+ A superfluous requirement about setting Content-Length + has been removed from the description of the OPTIONS method. + (Section 9.3.7)¶

+

+ The normative requirement to use the "message/http" media type in + TRACE responses has been removed. + (Section 9.3.8)¶

+

+ List-based grammar for Expect has been restored for compatibility with + RFC 2616. + (Section 10.1.1)¶

+

+ Accept and Accept-Encoding are allowed in response + messages; the latter was introduced by [RFC7694]. + (Section 12.3)¶

+

+ "Accept Parameters" (accept-params and accept-ext ABNF production) have + been removed from the definition of the Accept field. + (Section 12.5.1)¶

+

+ The Accept-Charset field is now deprecated. + (Section 12.5.2)¶

+

+ The semantics of "*" in the Vary header field when other + values are present was clarified. + (Section 12.5.5)¶

+

+ Range units are compared in a case-insensitive fashion. + (Section 14.1)¶

+

+ The use of the Accept-Ranges field is not restricted to origin servers. + (Section 14.3)¶

+

+ The process of creating a redirected request has been clarified. + (Section 15.4)¶

+

+ Status code 308 (previously defined in [RFC7538]) + has been added so that it's defined closer to status codes 301, 302, and 307. + (Section 15.4.9)¶

+

+ Status code 421 (previously defined in + Section 9.1.2 of [RFC7540]) has been added because of its general + applicability. 421 is no longer defined as heuristically cacheable since + the response is specific to the connection (not the target resource). + (Section 15.5.20)¶

+

+ Status code 422 (previously defined in + Section 11.2 of [WEBDAV]) has been added because of its general + applicability. + (Section 15.5.21)¶

+
+
+
+
+

+B.4. Changes from RFC 7232 +

+

+ Previous revisions of HTTP imposed an arbitrary 60-second limit on the + determination of whether Last-Modified was a strong validator to guard + against the possibility that the Date and Last-Modified values are + generated from different clocks or at somewhat different times during the + preparation of the response. This specification has relaxed that to allow + reasonable discretion. + (Section 8.8.2.2)¶

+

+ An edge-case requirement on If-Match and If-Unmodified-Since + has been removed that required a validator not to be sent in a 2xx + response if validation fails because the change request has already + been applied. + (Sections 13.1.1 and + 13.1.4)¶

+

+ The fact that If-Unmodified-Since does not apply to a resource without a + concept of modification time has been clarified. + (Section 13.1.4)¶

+

+ Preconditions can now be evaluated before the request content is processed + rather than waiting until the response would otherwise be successful. + (Section 13.2)¶

+
+
+
+
+

+B.5. Changes from RFC 7233 +

+

+ Refactored the range-unit and ranges-specifier grammars to simplify + and reduce artificial distinctions between bytes and other + (extension) range units, removing the overlapping grammar of + other-range-unit by defining range units generically as a token and + placing extensions within the scope of a range-spec (other-range). + This disambiguates the role of list syntax (commas) in all range sets, + including extension range units, for indicating a range-set of more than + one range. Moving the extension grammar into range specifiers also allows + protocol specific to byte ranges to be specified separately.¶

+

+ It is now possible to define Range handling on extension methods. + (Section 14.2)¶

+

+ Described use of the Content-Range header field + (Section 14.4) as a request modifier to perform a + partial PUT. + (Section 14.5)¶

+
+
+
+
+

+B.6. Changes from RFC 7235 +

+

+ None.¶

+
+
+
+
+

+B.7. Changes from RFC 7538 +

+

+ None.¶

+
+
+
+
+

+B.8. Changes from RFC 7615 +

+

+ None.¶

+
+
+
+
+

+B.9. Changes from RFC 7694 +

+

+ This specification includes the extension defined in [RFC7694] + but leaves out examples and deployment considerations.¶

+
+
+
+
+
+
+

+Acknowledgements +

+

+ Aside from the current editors, the following individuals deserve special + recognition for their contributions to early aspects of HTTP and its + core specifications: + Marc Andreessen, Tim Berners-Lee, Robert Cailliau, Daniel W. Connolly, + Bob Denny, John Franks, Jim Gettys, + Jean-François Groff, + Phillip M. Hallam-Baker, + Koen Holtman, Jeffery L. Hostetler, Shel Kaphan, + Dave Kristol, Yves Lafon, Scott D. Lawrence, + Paul J. Leach, Håkon W. Lie, + Ari Luotonen, Larry Masinter, Rob McCool, + Jeffrey C. Mogul, Lou Montulli, + David Morris, Henrik Frystyk Nielsen, Dave Raggett, Eric Rescorla, + Tony Sanders, Lawrence C. Stewart, + Marc VanHeyningen, and Steve Zilles.¶

+

+ This document builds on the many contributions + that went into past specifications of HTTP, including + [HTTP/1.0], + [RFC2068], + [RFC2145], + [RFC2616], + [RFC2617], + [RFC2818], + [RFC7230], + [RFC7231], + [RFC7232], + [RFC7233], + [RFC7234], and + [RFC7235]. + The acknowledgements within those documents still apply.¶

+

+ Since 2014, the following contributors have helped improve this + specification by reporting bugs, asking smart questions, drafting or + reviewing text, and evaluating issues:¶

+

+ Alan Egerton, + Alex Rousskov, + Amichai Rothman, + Amos Jeffries, + Anders Kaseorg, + Andreas Gebhardt, + Anne van Kesteren, + Armin Abfalterer, + Aron Duby, + Asanka Herath, + Asbjørn Ulsberg, + Asta Olofsson, + Attila Gulyas, + Austin Wright, + Barry Pollard, + Ben Burkert, + Benjamin Kaduk, + Björn Höhrmann, + Brad Fitzpatrick, + Chris Pacejo, + Colin Bendell, + Cory Benfield, + Cory Nelson, + Daisuke Miyakawa, + Dale Worley, + Daniel Stenberg, + Danil Suits, + David Benjamin, + David Matson, + David Schinazi, + Дилян Палаузов (Dilyan Palauzov), + Eric Anderson, + Eric Rescorla, + Éric Vyncke, + Erik Kline, + Erwin Pe, + Etan Kissling, + Evert Pot, + Evgeny Vrublevsky, + Florian Best, + Francesca Palombini, + Igor Lubashev, + James Callahan, + James Peach, + Jeffrey Yasskin, + Kalin Gyokov, + Kannan Goundan, + 奥 一穂 (Kazuho Oku), + Ken Murchison, + Krzysztof Maczyński, + Lars Eggert, + Lucas Pardue, + Martin Duke, + Martin Dürst, + Martin Thomson, + Martynas Jusevičius, + Matt Menke, + Matthias Pigulla, + Mattias Grenfeldt, + Michael Osipov, + Mike Bishop, + Mike Pennisi, + Mike Taylor, + Mike West, + Mohit Sethi, + Murray Kucherawy, + Nathaniel J. Smith, + Nicholas Hurley, + Nikita Prokhorov, + Patrick McManus, + Piotr Sikora, + Poul-Henning Kamp, + Rick van Rein, + Robert Wilton, + Roberto Polli, + Roman Danyliw, + Samuel Williams, + Semyon Kholodnov, + Simon Pieters, + Simon Schüppel, + Stefan Eissing, + Taylor Hunt, + Todd Greer, + Tommy Pauly, + Vasiliy Faronov, + Vladimir Lashchev, + Wenbo Zhu, + William A. Rowe Jr., + Willy Tarreau, + Xingwei Liu, + Yishuai Li, and + Zaheduzzaman Sarker.¶

+
+
+
+

+Index +

+
+

+ 1 + 2 + 3 + 4 + 5 + A + B + C + D + E + F + G + H + I + L + M + N + O + P + R + S + T + U + V + W + X¶

+
+ +
+
+
+

+Authors' Addresses +

+
+
Roy T. Fielding (editor)
+
Adobe
+
345 Park Ave
San Jose, CA 95110
+
United States of America
+ +
+URI: +https://roy.gbiv.com/ +
+
+
+
Mark Nottingham (editor)
+
Fastly
+
Prahran
+
Australia
+ +
+URI: +https://www.mnot.net/ +
+
+
+
Julian Reschke (editor)
+
greenbytes GmbH
+
Hafenweg 16
48155 Münster
+
Germany
+ +
+URI: +https://greenbytes.de/tech/webdav/ +
+
+
+
+ + + + diff --git a/skills/structured-extraction/fixtures/rfc9110-http-semantics.yaml b/skills/structured-extraction/fixtures/rfc9110-http-semantics.yaml new file mode 100644 index 000000000..040ab424b --- /dev/null +++ b/skills/structured-extraction/fixtures/rfc9110-http-semantics.yaml @@ -0,0 +1,21 @@ +name: rfc9110-http-semantics +lane: deterministic +target: + kind: skill + ref: . +runner: extract +inputs: + input_path: "fixtures/rfc9110-http-semantics.html" + schema_path: "schemas/extraction.schema.json" + source_url: "https://www.rfc-editor.org/rfc/rfc9110.html" + content_type: "text/html" + max_items: 18 +expect: + status: success + outputs: + structured_extraction_result: + matches_packet: runx.structured_extraction.result.v1 +metadata: + public_skill: structured-extraction + source_case: rfc9110-http-semantics + source: p-40c7eb6c41 diff --git a/skills/structured-extraction/package.json b/skills/structured-extraction/package.json new file mode 100644 index 000000000..e2ac61f46 --- /dev/null +++ b/skills/structured-extraction/package.json @@ -0,0 +1,7 @@ +{ + "name": "@runxhq/structured-extraction", + "version": "0.1.0", + "private": true, + "type": "module", + "description": "runx governed skill for deterministic structured extraction from messy HTML or text." +} diff --git a/skills/structured-extraction/schemas/extraction.schema.json b/skills/structured-extraction/schemas/extraction.schema.json new file mode 100644 index 000000000..d1501afe7 --- /dev/null +++ b/skills/structured-extraction/schemas/extraction.schema.json @@ -0,0 +1,277 @@ +{ + "$schema": "https://json-schema.org/draft/2020-12/schema", + "$id": "runx.structured_extraction.result.v1", + "type": "object", + "additionalProperties": false, + "required": [ + "schema", + "source", + "extraction", + "validation", + "provenance" + ], + "properties": { + "schema": { + "const": "runx.structured_extraction.result.v1" + }, + "source": { + "type": "object", + "additionalProperties": false, + "required": [ + "url", + "content_type", + "input_path", + "input_sha256", + "input_bytes" + ], + "properties": { + "url": { + "type": "string", + "minLength": 1 + }, + "content_type": { + "enum": [ + "text/html", + "text/plain" + ] + }, + "input_path": { + "type": "string", + "minLength": 1 + }, + "input_sha256": { + "type": "string", + "pattern": "^sha256:[0-9a-f]{64}$" + }, + "input_bytes": { + "type": "integer", + "minimum": 1 + } + } + }, + "extraction": { + "type": "object", + "additionalProperties": false, + "required": [ + "title", + "summary", + "items" + ], + "properties": { + "title": { + "type": "string", + "minLength": 1 + }, + "summary": { + "type": "object", + "additionalProperties": false, + "required": [ + "item_count", + "heading_count", + "term_count", + "paragraph_count", + "text_chars" + ], + "properties": { + "item_count": { + "type": "integer", + "minimum": 1 + }, + "heading_count": { + "type": "integer", + "minimum": 0 + }, + "term_count": { + "type": "integer", + "minimum": 0 + }, + "paragraph_count": { + "type": "integer", + "minimum": 0 + }, + "text_chars": { + "type": "integer", + "minimum": 1 + } + } + }, + "items": { + "type": "array", + "minItems": 8, + "items": { + "oneOf": [ + { + "$ref": "#/$defs/heading_item" + }, + { + "$ref": "#/$defs/paragraph_item" + }, + { + "$ref": "#/$defs/term_item" + } + ] + } + } + } + }, + "validation": { + "type": "object", + "additionalProperties": false, + "required": [ + "schema_id", + "schema_sha256", + "valid", + "engine", + "checks" + ], + "properties": { + "schema_id": { + "const": "runx.structured_extraction.result.v1" + }, + "schema_sha256": { + "type": "string", + "pattern": "^sha256:[0-9a-f]{64}$" + }, + "valid": { + "type": "boolean" + }, + "engine": { + "const": "native-json-schema-subset-v1" + }, + "checks": { + "type": "array", + "minItems": 4, + "items": { + "type": "object", + "additionalProperties": false, + "required": [ + "name", + "passed", + "detail" + ], + "properties": { + "name": { + "type": "string", + "minLength": 1 + }, + "passed": { + "type": "boolean" + }, + "detail": { + "type": "string" + } + } + } + } + } + }, + "provenance": { + "type": "object", + "additionalProperties": false, + "required": [ + "mode", + "tool_version", + "source_kind", + "output_payload_sha256" + ], + "properties": { + "mode": { + "const": "fixture" + }, + "tool_version": { + "type": "string", + "minLength": 1 + }, + "source_kind": { + "const": "real_public_document" + }, + "output_payload_sha256": { + "type": "string", + "pattern": "^sha256:[0-9a-f]{64}$" + } + } + }, + "artifacts": { + "type": "array", + "minItems": 3 + }, + "signal": { + "type": "object" + } + }, + "$defs": { + "heading_item": { + "type": "object", + "additionalProperties": false, + "required": [ + "kind", + "level", + "text", + "anchor" + ], + "properties": { + "kind": { + "const": "heading" + }, + "level": { + "type": "integer", + "minimum": 1, + "maximum": 4 + }, + "text": { + "type": "string", + "minLength": 3 + }, + "anchor": { + "anyOf": [ + { + "type": "string" + }, + { + "type": "null" + } + ] + } + } + }, + "paragraph_item": { + "type": "object", + "additionalProperties": false, + "required": [ + "kind", + "text" + ], + "properties": { + "kind": { + "const": "paragraph" + }, + "text": { + "type": "string", + "minLength": 80 + } + } + }, + "term_item": { + "type": "object", + "additionalProperties": false, + "required": [ + "kind", + "text", + "context" + ], + "properties": { + "kind": { + "const": "term" + }, + "text": { + "type": "string", + "minLength": 2 + }, + "context": { + "type": "string", + "minLength": 10 + } + } + } + } +} diff --git a/skills/structured-extraction/tools/structured/extract/fixtures/rfc9110.yaml b/skills/structured-extraction/tools/structured/extract/fixtures/rfc9110.yaml new file mode 100644 index 000000000..3ea6a73ea --- /dev/null +++ b/skills/structured-extraction/tools/structured/extract/fixtures/rfc9110.yaml @@ -0,0 +1,27 @@ +name: structured-extract-rfc9110 +lane: deterministic +target: + kind: tool + ref: structured.extract +inputs: + input_path: fixtures/rfc9110-http-semantics.html + schema_path: schemas/extraction.schema.json + source_url: https://www.rfc-editor.org/rfc/rfc9110.html + content_type: text/html + max_items: 18 +expect: + status: success + output: + matches_packet: runx.structured_extraction.result.v1 + subset: + schema: runx.structured_extraction.result.v1 + source: + input_path: fixtures/rfc9110-http-semantics.html + url: https://www.rfc-editor.org/rfc/rfc9110.html + content_type: text/html + validation: + schema_id: runx.structured_extraction.result.v1 + valid: true + provenance: + mode: fixture + source_kind: real_public_document diff --git a/skills/structured-extraction/tools/structured/extract/manifest.json b/skills/structured-extraction/tools/structured/extract/manifest.json new file mode 100644 index 000000000..0b3cc120e --- /dev/null +++ b/skills/structured-extraction/tools/structured/extract/manifest.json @@ -0,0 +1,65 @@ +{ + "schema": "runx.tool.manifest.v1", + "name": "structured.extract", + "version": "0.1.0", + "description": "Extract schema-validated JSON from a messy HTML or text fixture and emit digest-bound artifact references.", + "source": { + "type": "cli-tool", + "command": "node", + "args": [ + "./run.mjs" + ] + }, + "inputs": { + "input_path": { + "type": "string", + "required": true, + "description": "Package-relative messy HTML or text fixture." + }, + "schema_path": { + "type": "string", + "required": true, + "description": "Package-relative JSON Schema path." + }, + "source_url": { + "type": "string", + "required": true, + "description": "Canonical source URL for the fixture bytes." + }, + "content_type": { + "type": "string", + "required": false, + "default": "text/html", + "description": "Input content type." + }, + "max_items": { + "type": "number", + "required": false, + "default": 20, + "description": "Maximum extracted items to include." + } + }, + "scopes": [ + "runx:fixture:read", + "runx:schema:read" + ], + "runx": { + "artifacts": { + "named_emits": { + "structured_extraction_result": "runx.structured_extraction.result.v1" + }, + "wrap_as": "structured_extraction_result" + } + }, + "runtime": { + "command": "node", + "args": [ + "./run.mjs" + ] + }, + "output": { + "packet": "runx.structured_extraction.result.v1", + "wrap_as": "structured_extraction_result" + }, + "toolkit_version": "0.1.4" +} diff --git a/skills/structured-extraction/tools/structured/extract/run.mjs b/skills/structured-extraction/tools/structured/extract/run.mjs new file mode 100644 index 000000000..1ab8cef2f --- /dev/null +++ b/skills/structured-extraction/tools/structured/extract/run.mjs @@ -0,0 +1,415 @@ +import fs from "node:fs"; +import path from "node:path"; +import crypto from "node:crypto"; +import { fileURLToPath } from "node:url"; + +const __dirname = path.dirname(fileURLToPath(import.meta.url)); +const TOOL_VERSION = "0.1.0"; + +function readInputs() { + const raw = process.env.RUNX_INPUTS_PATH + ? fs.readFileSync(process.env.RUNX_INPUTS_PATH, "utf8") + : (process.env.RUNX_INPUTS_JSON || "{}"); + return JSON.parse(raw); +} + +function packageRoot() { + return path.resolve(__dirname, "../../.."); +} + +function resolveInsidePackage(relativePath, label) { + const root = packageRoot(); + const resolved = path.resolve(root, String(relativePath || "")); + if (!resolved.startsWith(root + path.sep) && resolved !== root) { + throw new Error(`${label} escapes the skill package`); + } + return resolved; +} + +function sha256Bytes(bytes) { + return `sha256:${crypto.createHash("sha256").update(bytes).digest("hex")}`; +} + +function sha256Text(text) { + return sha256Bytes(Buffer.from(text, "utf8")); +} + +function canonicalJson(value) { + if (value === null || typeof value !== "object") { + return JSON.stringify(value); + } + if (Array.isArray(value)) { + return `[${value.map(canonicalJson).join(",")}]`; + } + return `{${Object.keys(value).sort().map((key) => `${JSON.stringify(key)}:${canonicalJson(value[key])}`).join(",")}}`; +} + +function schemaTypeOf(value) { + if (value === null) { + return "null"; + } + if (Array.isArray(value)) { + return "array"; + } + if (Number.isInteger(value)) { + return "integer"; + } + return typeof value; +} + +function resolveSchemaRef(ref, rootSchema) { + if (!ref.startsWith("#/")) { + throw new Error(`unsupported schema ref: ${ref}`); + } + return ref + .slice(2) + .split("/") + .reduce((node, part) => { + const key = part.replace(/~1/g, "/").replace(/~0/g, "~"); + if (!node || typeof node !== "object" || !(key in node)) { + throw new Error(`unresolved schema ref: ${ref}`); + } + return node[key]; + }, rootSchema); +} + +function validateSchemaValue(value, schemaNode, rootSchema, pointer = "$") { + if (!schemaNode || typeof schemaNode !== "object") { + return []; + } + if (schemaNode.$ref) { + return validateSchemaValue(value, resolveSchemaRef(schemaNode.$ref, rootSchema), rootSchema, pointer); + } + if (Array.isArray(schemaNode.anyOf)) { + const variants = schemaNode.anyOf.map((variant) => validateSchemaValue(value, variant, rootSchema, pointer)); + if (variants.some((errors) => errors.length === 0)) { + return []; + } + return [`${pointer} did not match anyOf: ${variants.map((errors) => errors[0]).filter(Boolean).join("; ")}`]; + } + if (Array.isArray(schemaNode.oneOf)) { + const matches = schemaNode.oneOf + .map((variant) => validateSchemaValue(value, variant, rootSchema, pointer)) + .filter((errors) => errors.length === 0); + return matches.length === 1 ? [] : [`${pointer} matched ${matches.length} oneOf variants`]; + } + + const errors = []; + const actualType = schemaTypeOf(value); + if (schemaNode.type) { + const allowedTypes = Array.isArray(schemaNode.type) ? schemaNode.type : [schemaNode.type]; + if (!allowedTypes.includes(actualType)) { + errors.push(`${pointer} expected ${allowedTypes.join("|")}, got ${actualType}`); + return errors; + } + } + if ("const" in schemaNode && value !== schemaNode.const) { + errors.push(`${pointer} expected const ${JSON.stringify(schemaNode.const)}`); + } + if (Array.isArray(schemaNode.enum) && !schemaNode.enum.includes(value)) { + errors.push(`${pointer} expected one of ${schemaNode.enum.map((entry) => JSON.stringify(entry)).join(", ")}`); + } + if (typeof value === "string") { + if (schemaNode.minLength !== undefined && value.length < schemaNode.minLength) { + errors.push(`${pointer} length ${value.length} < ${schemaNode.minLength}`); + } + if (schemaNode.pattern && !(new RegExp(schemaNode.pattern).test(value))) { + errors.push(`${pointer} does not match ${schemaNode.pattern}`); + } + } + if (typeof value === "number") { + if (schemaNode.minimum !== undefined && value < schemaNode.minimum) { + errors.push(`${pointer} ${value} < ${schemaNode.minimum}`); + } + if (schemaNode.maximum !== undefined && value > schemaNode.maximum) { + errors.push(`${pointer} ${value} > ${schemaNode.maximum}`); + } + } + if (Array.isArray(value)) { + if (schemaNode.minItems !== undefined && value.length < schemaNode.minItems) { + errors.push(`${pointer} has ${value.length} item(s), expected at least ${schemaNode.minItems}`); + } + if (schemaNode.maxItems !== undefined && value.length > schemaNode.maxItems) { + errors.push(`${pointer} has ${value.length} item(s), expected at most ${schemaNode.maxItems}`); + } + if (schemaNode.items) { + value.forEach((item, index) => { + errors.push(...validateSchemaValue(item, schemaNode.items, rootSchema, `${pointer}[${index}]`)); + }); + } + } + if (value && typeof value === "object" && !Array.isArray(value)) { + const properties = schemaNode.properties || {}; + const required = Array.isArray(schemaNode.required) ? schemaNode.required : []; + for (const field of required) { + if (!(field in value)) { + errors.push(`${pointer}.${field} is required`); + } + } + for (const [field, fieldValue] of Object.entries(value)) { + if (field in properties) { + errors.push(...validateSchemaValue(fieldValue, properties[field], rootSchema, `${pointer}.${field}`)); + } else if (schemaNode.additionalProperties === false) { + errors.push(`${pointer}.${field} is not allowed`); + } + } + } + return errors; +} + +function decodeEntities(text) { + return text + .replace(/ /g, " ") + .replace(/&/g, "&") + .replace(/</g, "<") + .replace(/>/g, ">") + .replace(/"/g, "\"") + .replace(/'/g, "'") + .replace(/&#x([0-9a-f]+);/gi, (_, hex) => String.fromCodePoint(Number.parseInt(hex, 16))) + .replace(/&#([0-9]+);/g, (_, dec) => String.fromCodePoint(Number.parseInt(dec, 10))); +} + +function stripTags(html) { + return decodeEntities(html) + .replace(//gi, " ") + .replace(//gi, " ") + .replace(/<[^>]+>/g, " ") + .replace(/\s+/g, " ") + .trim(); +} + +function extractTitle(raw, contentType) { + if (contentType === "text/html") { + const title = raw.match(/]*>([\s\S]*?)<\/title>/i); + if (title) { + return stripTags(title[1]); + } + const h1 = raw.match(/]*>([\s\S]*?)<\/h1>/i); + if (h1) { + return stripTags(h1[1]); + } + } + return raw.split(/\r?\n/).map((line) => line.trim()).find(Boolean) || "Untitled input"; +} + +function extractHeadings(raw, maxItems) { + const items = []; + const headingPattern = /]*)>([\s\S]*?)<\/h\1>/gi; + let match; + while ((match = headingPattern.exec(raw)) && items.length < maxItems) { + const attrs = match[2] || ""; + const id = attrs.match(/\sid=["']([^"']+)["']/i)?.[1] || null; + const text = stripTags(match[3]); + if (text.length >= 3) { + items.push({ + kind: "heading", + level: Number(match[1]), + text, + anchor: id, + }); + } + } + return items; +} + +function extractHttpTokens(text, maxItems) { + const tokenPattern = /\b(GET|HEAD|POST|PUT|DELETE|CONNECT|OPTIONS|TRACE|PATCH|HTTP\/[0-9.]+|Content-Type|Content-Length|Cache-Control|Authorization|Location|ETag|Accept)\b/g; + const seen = new Set(); + const items = []; + let match; + while ((match = tokenPattern.exec(text)) && items.length < maxItems) { + const token = match[1]; + if (seen.has(token)) { + continue; + } + seen.add(token); + const start = Math.max(0, match.index - 90); + const end = Math.min(text.length, match.index + token.length + 90); + items.push({ + kind: "term", + text: token, + context: text.slice(start, end).replace(/\s+/g, " ").trim(), + }); + } + return items; +} + +function extractParagraphs(raw, maxItems) { + const paragraphs = []; + const pattern = /]*>([\s\S]*?)<\/p>/gi; + let match; + while ((match = pattern.exec(raw)) && paragraphs.length < maxItems) { + const text = stripTags(match[1]); + if (text.length > 80) { + paragraphs.push({ + kind: "paragraph", + text: text.slice(0, 360), + }); + } + } + return paragraphs; +} + +function validatePacket(packet, schema) { + const counts = { + heading: packet.extraction.items.filter((item) => item.kind === "heading").length, + term: packet.extraction.items.filter((item) => item.kind === "term").length, + paragraph: packet.extraction.items.filter((item) => item.kind === "paragraph").length, + }; + const checks = [ + { name: "has_source_digest", passed: /^sha256:[0-9a-f]{64}$/.test(packet.source.input_sha256), detail: packet.source.input_sha256 }, + { name: "has_schema_digest", passed: /^sha256:[0-9a-f]{64}$/.test(packet.validation.schema_sha256), detail: packet.validation.schema_sha256 }, + { name: "has_minimum_items", passed: Array.isArray(packet.extraction.items) && packet.extraction.items.length >= 8, detail: String(packet.extraction.items.length) }, + { + name: "summary_counts_match_items", + passed: + packet.extraction.summary.item_count === packet.extraction.items.length && + packet.extraction.summary.heading_count === counts.heading && + packet.extraction.summary.term_count === counts.term && + packet.extraction.summary.paragraph_count === counts.paragraph, + detail: JSON.stringify(counts), + }, + ]; + packet.validation.checks = checks; + packet.validation.valid = checks.every((check) => check.passed); + const schemaErrors = validateSchemaValue(packet, schema, schema); + checks.push({ + name: "json_schema_validation", + passed: schemaErrors.length === 0, + detail: schemaErrors.length === 0 ? "output validates against schemas/extraction.schema.json" : schemaErrors.slice(0, 6).join("; "), + }); + packet.validation.valid = checks.every((check) => check.passed); + return { + valid: packet.validation.valid, + checks, + schemaErrors, + }; +} + +function main() { + const inputs = readInputs(); + const inputPath = resolveInsidePackage(inputs.input_path, "input_path"); + const schemaPath = resolveInsidePackage(inputs.schema_path, "schema_path"); + const contentType = String(inputs.content_type || "text/html"); + const maxItems = Math.max(8, Math.min(Number(inputs.max_items || 20), 60)); + const sourceUrl = String(inputs.source_url || "").trim(); + if (!sourceUrl) { + throw new Error("source_url is required"); + } + + const inputBytes = fs.readFileSync(inputPath); + const raw = inputBytes.toString("utf8"); + const schemaBytes = fs.readFileSync(schemaPath); + const schema = JSON.parse(schemaBytes.toString("utf8")); + const plainText = contentType === "text/html" ? stripTags(raw) : raw.replace(/\s+/g, " ").trim(); + + const headings = contentType === "text/html" ? extractHeadings(raw, Math.ceil(maxItems / 2)) : []; + const paragraphs = contentType === "text/html" ? extractParagraphs(raw, 3) : []; + const terms = extractHttpTokens(plainText, maxItems); + const items = [...headings, ...paragraphs, ...terms].slice(0, maxItems); + + const extraction = { + title: extractTitle(raw, contentType), + summary: { + item_count: items.length, + heading_count: items.filter((item) => item.kind === "heading").length, + term_count: items.filter((item) => item.kind === "term").length, + paragraph_count: items.filter((item) => item.kind === "paragraph").length, + text_chars: plainText.length, + }, + items, + }; + + const packet = { + schema: "runx.structured_extraction.result.v1", + source: { + url: sourceUrl, + content_type: contentType, + input_path: String(inputs.input_path), + input_sha256: sha256Bytes(inputBytes), + input_bytes: inputBytes.length, + }, + extraction, + validation: { + schema_id: schema.$id || "runx.structured_extraction.result.v1", + schema_sha256: sha256Bytes(schemaBytes), + valid: false, + engine: "native-json-schema-subset-v1", + checks: [], + }, + provenance: { + mode: "fixture", + tool_version: TOOL_VERSION, + source_kind: "real_public_document", + output_payload_sha256: null, + }, + }; + + const outputPayload = { + source: packet.source, + extraction: packet.extraction, + validation_schema_id: packet.validation.schema_id, + }; + packet.provenance.output_payload_sha256 = sha256Text(canonicalJson(outputPayload)); + const validation = validatePacket(packet, schema); + packet.validation.valid = validation.valid; + packet.validation.checks = validation.checks; + const finalSchemaErrors = validateSchemaValue(packet, schema, schema); + if (finalSchemaErrors.length > 0) { + packet.validation.valid = false; + packet.validation.checks.push({ + name: "final_json_schema_validation", + passed: false, + detail: finalSchemaErrors.slice(0, 6).join("; "), + }); + } + if (!packet.validation.valid) { + throw new Error(`schema validation failed: ${JSON.stringify(packet.validation.checks)}`); + } + + packet.artifacts = [ + { + id: packet.source.input_sha256, + artifact_id: packet.source.input_sha256, + type: "input_fixture", + artifact_type: "input_fixture", + label: "RFC 9110 HTML fixture", + source_url: sourceUrl, + byte_count: inputBytes.length, + }, + { + id: packet.validation.schema_sha256, + artifact_id: packet.validation.schema_sha256, + type: "json_schema", + artifact_type: "json_schema", + label: packet.validation.schema_id, + }, + { + id: packet.provenance.output_payload_sha256, + artifact_id: packet.provenance.output_payload_sha256, + type: "validated_output", + artifact_type: "validated_output", + label: "Structured extraction JSON payload", + }, + ]; + packet.signal = { + signal_id: `structured-extraction:${packet.source.input_sha256}:${packet.provenance.output_payload_sha256}`, + source_events: [ + { + provider: "rfc-editor", + source_locator: sourceUrl, + title: "RFC 9110 HTTP Semantics HTML", + }, + ], + artifacts: packet.artifacts, + }; + + process.stdout.write(JSON.stringify(packet)); +} + +try { + main(); +} catch (error) { + process.stderr.write(`${JSON.stringify({ error: { message: error.message } })}\n`); + process.exitCode = 1; +} diff --git a/skills/support-triage-reply/SKILL.md b/skills/support-triage-reply/SKILL.md new file mode 100644 index 000000000..ec7b74de7 --- /dev/null +++ b/skills/support-triage-reply/SKILL.md @@ -0,0 +1,83 @@ +--- +name: support-triage-reply +version: 0.1.0 +description: Classify a bounded support request, choose the safe next path, and draft a customer-ready reply only when a human-gated send is appropriate. +source: + type: cli-tool + command: node + args: + - run.mjs +links: + source: https://github.com/runxhq/runx/tree/main/skills/support-triage-reply +runx: + category: ops + input_resolution: + required: + - support_request +--- + +# Support Triage Reply + +Classify one bounded support request and return a support-safe decision packet. +The skill is designed for day-to-day operator work where support, product, and +engineering signals arrive together, but a customer send must remain a separate +human-approved action. + +This skill never sends email, posts to Slack, creates issues, mutates accounts, +or touches billing. It returns a draft and a gated send proposal only when the +request is safe to answer from the supplied context. + +## Inputs + +- `support_request`: object with `subject`, `body`, optional `customer_name`, + optional `customer_email`, optional `source`, and optional `refs`. +- `policy`: optional object with `product_name`, `support_signature`, + `safe_reply_topics`, and `escalation_contacts`. + +## Output + +The runner emits these top-level fields: + +- `classification`: `how_to`, `billing`, `account_access`, `bug`, `abuse`, or + `unknown`. +- `severity`: `low`, `medium`, `high`, or `critical`. +- `confidence`: number from 0 to 1. +- `recommended_path`: one of `reply_draft`, `request_info`, + `engineering_intake`, `billing_review`, `account_review`, `abuse_review`, or + `manual_review`. +- `evidence`: object with matched signals, missing context, source summary, and + taxonomy coverage. +- `draft_email`: object with `proposed`, `subject`, and `body`. When a reply is + not safe from the supplied packet, `proposed` is `false` and `reason` explains + the blocker. +- `send_gate`: object whose `status` is always `requires_human_approval`. + +## Decision Rules + +Prefer safety over completeness: + +- `how_to`: draft a clear support email when the request is answerable from the + supplied text or common product-safe instructions. +- `billing`: route to billing review unless the supplied context already names + a public, non-account-specific policy. +- `account_access`: route to account review. Do not ask for passwords, recovery + secrets, or private tokens. +- `bug`: route to engineering intake when the report includes a failure mode, + product surface, or reproduction clue. +- `abuse`: route to abuse review and do not draft a customer-facing answer. +- `unknown`: request more information or manual review. Do not invent a fix. + +Customer-facing copy must be specific, calm, and sendable. It should include a +greeting, acknowledge the actual request, state the answer or next step, and end +with the configured support signature. Avoid filler, unsupported promises, and +fake certainty. + +## Safety Bar + +- No customer send occurs inside this skill. +- No private credentials, billing records, account identifiers, or inbox state + are required. +- A draft is a proposal, not permission. The caller must use a separate + governed send lane to deliver any message. +- When confidence is low or private account state is needed, return + `manual_review` or a review-specific route. diff --git a/skills/support-triage-reply/X.yaml b/skills/support-triage-reply/X.yaml new file mode 100644 index 000000000..f71b51bce --- /dev/null +++ b/skills/support-triage-reply/X.yaml @@ -0,0 +1,90 @@ +skill: support-triage-reply +version: "0.1.0" + +catalog: + kind: skill + audience: operator + visibility: public + role: canonical + +harness: + cases: + - name: safe-how-to-reply-draft + runner: triage + inputs: + support_request: + customer_name: Mira + customer_email: mira@example.test + subject: How do I verify my sending domain? + body: I added the DNS records for my domain. What should I check next so emails can send safely? + source: fixture:safe-how-to + policy: + product_name: Nitrosend + support_signature: Nitrosend Support + expect: + status: sealed + receipt: + schema: runx.receipt.v1 + state: sealed + disposition: closed + reason_code: process_closed + + - name: account-access-escalates-without-draft + runner: triage + inputs: + support_request: + customer_name: Theo + customer_email: theo@example.test + subject: I cannot access my account + body: My login is blocked and I need someone to reset access for the team owner. + source: fixture:account-access + policy: + product_name: Nitrosend + support_signature: Nitrosend Support + expect: + status: sealed + receipt: + schema: runx.receipt.v1 + state: sealed + disposition: closed + reason_code: process_closed + + - name: missing-request-fails + runner: triage + inputs: + support_request: {} + expect: + status: failure + receipt: + schema: runx.receipt.v1 + state: sealed + disposition: closed + reason_code: process_failed + +runners: + triage: + default: true + type: cli-tool + command: node + args: + - run.mjs + outputs: + classification: string + severity: string + confidence: number + recommended_path: string + evidence: object + draft_email: object + send_gate: object + artifacts: + wrap_as: support_triage_packet + packet: runx.support.triage_reply.v1 + inputs: + support_request: + type: json + required: true + description: Bounded support request packet. + policy: + type: json + required: false + description: Product-safe support policy and signature hints. diff --git a/skills/support-triage-reply/references/dogfood-receipt.json b/skills/support-triage-reply/references/dogfood-receipt.json new file mode 100644 index 000000000..6a5ad040c --- /dev/null +++ b/skills/support-triage-reply/references/dogfood-receipt.json @@ -0,0 +1 @@ +{"schema":"runx.receipt.v1","id":"sha256:8e87b9c4c18bf2e39c976059bb480bb6825400f15bd62886d1592a57c7357b9f","created_at":"2026-06-20T18:10:51.168Z","canonicalization":"runx.receipt.c14n.v1","issuer":{"type":"hosted","kid":"runx-demo-key","public_key_sha256":"sha256:3097e2dee2cb4a34b53840cdb705aed71067c36f68db0e0f559c3f3fa043315f"},"signature":{"alg":"Ed25519","value":"base64:f80xpAwGB_2ZauvoeYM6XwjBEe15t2qA6FdNPlLrJTlGNSQ17-oYGRW4fm3pWBijd5M1cxUhC1KKC0ZNpi4QDw"},"digest":"sha256:a1e586ce442e81b2be02322835c116b4b046be9a7518dff0317b072ea8ed9f0c","idempotency":{"intent_key":"sha256:run_triage_1af024d047dc-triage-intent","trigger_fingerprint":"sha256:run_triage_1af024d047dc-triage-trigger","content_hash":"sha256:run_triage_1af024d047dc-triage-content"},"subject":{"kind":"skill","ref":{"type":"harness","uri":"hrn_run_triage_1af024d047dc_triage"},"commitments":[]},"authority":{"actor_ref":{"type":"principal","uri":"runx:principal:local_runtime"},"grant_refs":[],"scope_refs":[],"authority_proof_refs":[],"attenuation":{"parent_authority_ref":null,"subset_proof":null},"terms":[],"enforcement":{"profile_hash":"sha256:runtime-skeleton-enforcement","redaction_refs":[],"setup_refs":[],"teardown_refs":[]}},"signals":[],"decisions":[{"decision_id":"dec_triage","choice":"open","inputs":{"signal_refs":[],"target_ref":null,"opportunity_refs":[],"selection_ref":null},"proposed_intent":{"purpose":"Open runtime node triage","legitimacy":"Local graph execution requested this node","success_criteria":[],"constraints":[],"derived_from":[]},"selected_act_id":"act_triage","selected_harness_ref":null,"justification":{"summary":"runtime graph planner selected this node","evidence_refs":[]},"closure":null,"artifact_refs":[]}],"acts":[{"id":"act_triage","form":"observation","intent":{"purpose":"Run graph step triage","legitimacy":"Runtime graph execution was admitted by the local harness","success_criteria":[{"criterion_id":"process_exit","statement":"cli-tool exits successfully","required":true}],"constraints":[],"derived_from":[]},"summary":"Executed graph step triage","criterion_bindings":[{"criterion_id":"process_exit","status":"verified","evidence_refs":[],"verification_refs":[],"summary":"cli-tool exited successfully"}],"source_refs":[],"target_refs":[],"artifact_refs":[],"closure":{"disposition":"closed","reason_code":"process_exit","summary":"cli-tool exited successfully","closed_at":"2026-06-20T18:10:51.168Z"}}],"seal":{"disposition":"closed","reason_code":"process_closed","summary":"cli-tool triage completed","closed_at":"2026-06-20T18:10:51.168Z","last_observed_at":"2026-06-20T18:10:51.168Z","criteria":[{"criterion_id":"process_exit","status":"verified","evidence_refs":[],"verification_refs":[],"summary":"cli-tool exited successfully"}]},"lineage":{"children":[],"sync":[]}} \ No newline at end of file diff --git a/skills/support-triage-reply/references/evidence.json b/skills/support-triage-reply/references/evidence.json new file mode 100644 index 000000000..b1dfb323e --- /dev/null +++ b/skills/support-triage-reply/references/evidence.json @@ -0,0 +1,208 @@ +{ + "schema": "runx.support_triage_reply.evidence.v1", + "summary": "Published and dogfooded a reusable public Runx support-triage-reply skill for real operator support work. The skill classifies one support request, chooses a safe next path, drafts a customer-ready email only when appropriate, and keeps any customer send behind an explicit human approval gate.", + "skill": { + "name": "support-triage-reply", + "published_ref": "godfood/support-triage-reply@sha-4887b7e3476f", + "owner": "godfood", + "publisher_principal": "user_godfood_2190fb7211ca", + "registry": "https://api.runx.ai", + "public_url": "https://runx.ai/x/godfood/support-triage-reply", + "source_url": "https://github.com/runxhq/runx/tree/main/skills/support-triage-reply", + "digest": "d9c8eda288ab87a3c48b3cd58152b00072eea54e8b882ac5b57601f3b29e3b63", + "profile_digest": "362867317f6299483d754cef105e73057d1fe83ad7f60bd9c3086a844641765e", + "trust_tier": "community" + }, + "boundary": { + "nitrosend_private_skill_reused": false, + "source_shape": "A generic public Runx skill inspired by the same support-ops category as Nitrosend private triage/intake skills.", + "private_semantics_included": false, + "sends_or_mutates": false, + "send_gate": "requires_human_approval" + }, + "checks": { + "local_harness": { + "status": "passed", + "case_count": 3, + "case_names": [ + "safe-how-to-reply-draft", + "account-access-escalates-without-draft", + "missing-request-fails" + ], + "receipt_ids": [ + "sha256:d864febac35254bcf49995b68e3c71a77444f5ad73c0ac7d05150fa3df72d4e4", + "sha256:29ed6188a804af189b6772855305f5da08ed399127ebae1120491d69c3e8f6af", + "sha256:69d5d7d9ae0330c9b40c5739d874e54add6cd94c81da693188d237af996d6dec" + ] + }, + "registry_readback": { + "status": "passed", + "publisher_handle": "godfood", + "markdown_contract_checked": true, + "runner_names": [ + "triage" + ] + }, + "clean_install": { + "status": "installed", + "command": "runx add godfood/support-triage-reply@sha-4887b7e3476f --registry https://api.runx.ai --installation-id godfood-support-triage-final3 --json" + }, + "dogfood_run": { + "status": "passed", + "command": "runx skill godfood/support-triage-reply@sha-4887b7e3476f --registry https://api.runx.ai --input support_request= --input policy= --receipts --json", + "receipt_id": "sha256:8e87b9c4c18bf2e39c976059bb480bb6825400f15bd62886d1592a57c7357b9f", + "receipt_file": "dogfood-receipt.json", + "verify_command": "runx verify --receipt dogfood-receipt.json --json", + "verify_note": "Verified with the public Ed25519 verification key for the demo signing key used by the local runtime.", + "verify_valid": true, + "classification": "how_to", + "severity": "low", + "confidence": 0.88, + "recommended_path": "reply_draft", + "send_gate_status": "requires_human_approval", + "draft_email": { + "body": "Hi Mira,\n\nThanks for the note. You asked about How do I verify my sending domain?.\n\nFor Nitrosend domain verification, check that the DNS records shown in the sending-domain setup are published exactly, then run the domain verification check again after DNS propagation. If a record still fails, compare the host/name and value fields character for character, including whether your DNS provider automatically appends the root domain.\n\nBefore sending, an operator should confirm the product state and any account-specific details. This draft has not been sent.\n\nThanks,\nNitrosend Support", + "proposed": true, + "recipient_hint": "customer_email_present", + "subject": "Re: How do I verify my sending domain?" + }, + "output_evidence": { + "matched_signals": [ + "how do i", + "what should", + "verify", + "dns", + "domain" + ], + "missing_context": [], + "private_data_required": false, + "send_side_effects": "none", + "source": "dogfood:operator-support", + "source_summary": "How do I verify my sending domain?", + "taxonomy_coverage": [ + "how_to", + "billing", + "account_access", + "bug", + "abuse", + "unknown" + ] + } + }, + "hosted_run_admission": { + "status": "accepted_pending_worker", + "run_id": "hr_7f1f591110f9458295eefe13f1d25e8b", + "app_url": "https://runx.ai/r/hr_7f1f591110f9458295eefe13f1d25e8b", + "inspect_url": "https://api.runx.ai/v1/runs/hr_7f1f591110f9458295eefe13f1d25e8b", + "observed_status": "pending", + "note": "Hosted control accepted the run under the godfood principal, but the worker did not drain it during the verification window. The execution proof for this delivery is the registry-resolved local run receipt." + } + }, + "credential_hygiene": { + "temporary_godfood_publish_credentials_revoked": true, + "temporary_godfood_run_credentials_revoked": true, + "live_godfood_self_serve_credentials_remaining": 0, + "tokens_in_artifacts": false + }, + "observations": [ + "runx --version returned runx-cli 0.6.6, satisfying the CLI floor for this bounty.", + "The published registry ref is godfood/support-triage-reply@sha-4887b7e3476f under the godfood authenticated namespace.", + "runx registry read godfood/support-triage-reply@sha-4887b7e3476f --registry https://api.runx.ai --json resolves the package metadata, digests, publisher, trust tier, and triage runner.", + "A clean install succeeded with runx add godfood/support-triage-reply@sha-4887b7e3476f --registry https://api.runx.ai --installation-id godfood-support-triage-final3 --json.", + "The local harness passed three cases: safe-how-to-reply-draft, account-access-escalates-without-draft, and missing-request-fails.", + "The registry publish gate ran the hosted harness successfully before accepting the package.", + "The dogfood command ran the published registry skill on a real support-domain input and produced receipt sha256:8e87b9c4c18bf2e39c976059bb480bb6825400f15bd62886d1592a57c7357b9f.", + "runx verify --receipt dogfood-receipt.json --json returned valid for the dogfood receipt with kid runx-demo-key and public key IVL40Zt5HSRFMkLhXy6rbLfP+ntqXtMAl5YOBpiB2xI=.", + "The dogfood output classified the request as how_to with severity low, confidence 0.88, recommended_path reply_draft, and send_gate.status requires_human_approval.", + "The classification taxonomy covers how_to, billing, account_access, bug, abuse, and unknown; unsafe account, billing, abuse, bug, and unknown paths do not automatically propose a customer send.", + "The dogfood output includes draft_email.body with a greeting, acknowledgement of the domain-verification request, concrete DNS verification steps, an explicit not-sent notice, and the configured support signoff." + ], + "observation_details": { + "runx_version": "runx-cli 0.6.6", + "publisher_owner": "godfood", + "package_name": "support-triage-reply", + "version": "sha-4887b7e3476f", + "registry_ref": "godfood/support-triage-reply@sha-4887b7e3476f", + "public_url": "https://runx.ai/x/godfood/support-triage-reply", + "source_url": "https://github.com/runxhq/runx/tree/main/skills/support-triage-reply", + "publish_method": "purpose-scoped godfood publish credential; credential revoked after publish", + "install_command": "runx add godfood/support-triage-reply@sha-4887b7e3476f --registry https://api.runx.ai --installation-id godfood-support-triage-final3 --json", + "dogfood_command": "runx skill godfood/support-triage-reply@sha-4887b7e3476f --registry https://api.runx.ai --input support_request= --input policy= --receipts --json", + "verify_command": "runx verify --receipt dogfood-receipt.json --json", + "verify_public_key_base64": "IVL40Zt5HSRFMkLhXy6rbLfP+ntqXtMAl5YOBpiB2xI=", + "harness_case_names": [ + "safe-how-to-reply-draft", + "account-access-escalates-without-draft", + "missing-request-fails" + ], + "local_harness_status": "passed", + "hosted_registry_harness_status": "passed as the registry publish gate", + "receipt_ref": "sha256:8e87b9c4c18bf2e39c976059bb480bb6825400f15bd62886d1592a57c7357b9f", + "runx_verify_verdict": "valid", + "classification_taxonomy_coverage": [ + "how_to", + "billing", + "account_access", + "bug", + "abuse", + "unknown" + ], + "safe_answer_case": "safe-how-to-reply-draft", + "escalation_case": "account-access-escalates-without-draft", + "how_to_install": "runx add godfood/support-triage-reply@sha-4887b7e3476f --registry https://api.runx.ai --installation-id ", + "how_to_run": "runx skill godfood/support-triage-reply@sha-4887b7e3476f --registry https://api.runx.ai --input support_request= --input policy= --json", + "how_to_verify": "Download dogfood-receipt.json from source, set RUNX_RECEIPT_VERIFY_KID=runx-demo-key and RUNX_RECEIPT_VERIFY_ED25519_PUBLIC_KEY_BASE64 to the public key in this evidence file, then run runx verify --receipt dogfood-receipt.json --json.", + "dogfood_draft_subject": "Re: How do I verify my sending domain?", + "dogfood_draft_body": "Hi Mira,\n\nThanks for the note. You asked about How do I verify my sending domain?.\n\nFor Nitrosend domain verification, check that the DNS records shown in the sending-domain setup are published exactly, then run the domain verification check again after DNS propagation. If a record still fails, compare the host/name and value fields character for character, including whether your DNS provider automatically appends the root domain.\n\nBefore sending, an operator should confirm the product state and any account-specific details. This draft has not been sent.\n\nThanks,\nNitrosend Support", + "dogfood_output": { + "classification": "how_to", + "confidence": 0.88, + "draft_email": { + "body": "Hi Mira,\n\nThanks for the note. You asked about How do I verify my sending domain?.\n\nFor Nitrosend domain verification, check that the DNS records shown in the sending-domain setup are published exactly, then run the domain verification check again after DNS propagation. If a record still fails, compare the host/name and value fields character for character, including whether your DNS provider automatically appends the root domain.\n\nBefore sending, an operator should confirm the product state and any account-specific details. This draft has not been sent.\n\nThanks,\nNitrosend Support", + "proposed": true, + "recipient_hint": "customer_email_present", + "subject": "Re: How do I verify my sending domain?" + }, + "evidence": { + "matched_signals": [ + "how do i", + "what should", + "verify", + "dns", + "domain" + ], + "missing_context": [], + "private_data_required": false, + "send_side_effects": "none", + "source": "dogfood:operator-support", + "source_summary": "How do I verify my sending domain?", + "taxonomy_coverage": [ + "how_to", + "billing", + "account_access", + "bug", + "abuse", + "unknown" + ] + }, + "recommended_path": "reply_draft", + "send_gate": { + "action": "send_customer_email", + "rationale": "The draft is customer-ready, but this skill never sends. A separate governed send lane must approve delivery.", + "status": "requires_human_approval" + }, + "severity": "low" + } + }, + "value_assessment": { + "real_operator_value": true, + "summary": "The skill handles a repeatable day-to-day support operation: classify one support request, select the safe lane, and draft a sendable reply only behind a human approval gate.", + "why_not_throwaway": "It is a reusable public Runx skill with a registry owner, source path, harness cases, typed outputs, safety gates, and receipt proof. It is not a one-off hosted webpage or private product-only wrapper.", + "review_bar": [ + "No automatic customer sends.", + "No private account or billing lookup is claimed.", + "Account, billing, abuse, bug, and unknown cases route away from unsafe customer replies.", + "The caller receives enough structured evidence to decide whether a human should send, escalate, or request more information." + ] + } +} diff --git a/skills/support-triage-reply/references/report.md b/skills/support-triage-reply/references/report.md new file mode 100644 index 000000000..dd34cd723 --- /dev/null +++ b/skills/support-triage-reply/references/report.md @@ -0,0 +1,127 @@ +# Support Triage Reply Dogfood Report + +## Result + +`support-triage-reply` is published as: + +- Registry ref: `godfood/support-triage-reply@` +- Public page: https://runx.ai/x/godfood/support-triage-reply +- Source: https://github.com/runxhq/runx/tree/main/skills/support-triage-reply +- Digest and profile digest: resolve with + `runx registry read godfood/support-triage-reply@ --registry https://api.runx.ai --json` +- Trust tier: `community` + +The skill is intentionally generic. Nitrosend has private support-ops skills +for triage/intake, but this public artifact does not include Nitrosend-private +policy, credentials, customer data, or mutation paths. + +## What It Does + +The skill accepts one bounded `support_request` and optional `policy`, then +returns: + +- `classification` +- `severity` +- `confidence` +- `recommended_path` +- `evidence` +- `draft_email` +- `send_gate` + +The skill never sends email, posts comments, mutates accounts, opens issues, or +touches billing. A reply draft is only a proposal; `send_gate.status` remains +`requires_human_approval`. + +## Verification + +Local harness: + +```sh +runx harness skills/support-triage-reply --receipt-dir "$tmp_receipts" --json +``` + +Result: passed, 3 cases: + +- `safe-how-to-reply-draft` +- `account-access-escalates-without-draft` +- `missing-request-fails` + +Clean install: + +```sh +runx add godfood/support-triage-reply@ \ + --registry https://api.runx.ai \ + --installation-id godfood-support-triage-final \ + --json +``` + +Result: installed. + +Dogfood execution: + +```sh +runx skill godfood/support-triage-reply@ \ + --registry https://api.runx.ai \ + --input 'support_request=' \ + --input 'policy=' \ + --receipts \ + --json +``` + +Output summary: + +- Receipt: see `evidence.json` and `dogfood-receipt.json` +- Classification: `how_to` +- Severity: `low` +- Confidence: `0.88` +- Recommended path: `reply_draft` +- Send gate: `requires_human_approval` + +Draft excerpt from the dogfood run: + +```text +Hi Mira, + +Thanks for the note. You asked about How do I verify my sending domain?. + +For Nitrosend domain verification, check that the DNS records shown in the sending-domain setup are published exactly, then run the domain verification check again after DNS propagation. If a record still fails, compare the host/name and value fields character for character, including whether your DNS provider automatically appends the root domain. + +Before sending, an operator should confirm the product state and any account-specific details. This draft has not been sent. + +Thanks, +Nitrosend Support +``` + +Receipt verification: + +```sh +runx verify --receipt dogfood-receipt.json --json +``` + +Result: valid. + +Hosted control admission was also exercised as `godfood`: + +- Run: `hr_7f1f591110f9458295eefe13f1d25e8b` +- URL: https://runx.ai/r/hr_7f1f591110f9458295eefe13f1d25e8b +- Observed state: `pending` + +The hosted worker did not drain during the verification window, so the execution +proof for this delivery is the registry-resolved local receipt above. Temporary +`godfood` publish/run credentials used for this dogfood pass were revoked after +use. + +## Review Value + +This is review-worthy because it converts a common, real operator workflow into +a reusable governed skill: + +- It distinguishes safe support replies from account, billing, abuse, bug, and + unknown cases. +- It returns explicit evidence and missing context instead of pretending to know + private account state. +- It gives a human a sendable draft when appropriate, but never finalizes a send. +- It is public, installable, source-backed, and receipt-backed. + +This is not a throwaway deployment. It is a reusable ops skill with a registry +entry, harness, typed runner, and public source. diff --git a/skills/support-triage-reply/run.mjs b/skills/support-triage-reply/run.mjs new file mode 100644 index 000000000..93d5e78ac --- /dev/null +++ b/skills/support-triage-reply/run.mjs @@ -0,0 +1,231 @@ +import fs from "node:fs"; + +const inputs = readInputs(); +const request = objectValue(inputs.support_request, "support_request"); +const policy = objectValue(inputs.policy ?? {}, "policy"); + +const subject = stringValue(request.subject); +const body = stringValue(request.body); + +if (!subject && !body) { + fail("support_request.subject or support_request.body is required"); +} + +const text = `${subject ?? ""}\n${body ?? ""}`; +const normalized = normalize(text); +const classification = classify(normalized); +const severity = classifySeverity(classification, normalized); +const confidence = confidenceFor(classification, normalized); +const recommendedPath = recommendedPathFor(classification, confidence); +const productName = stringValue(policy.product_name) ?? "the product"; +const signature = stringValue(policy.support_signature) ?? "Support"; +const customerName = firstName(stringValue(request.customer_name)); +const draftEmail = recommendedPath === "reply_draft" + ? buildDraftEmail({ request, subject, body, productName, signature, customerName }) + : { + proposed: false, + subject: null, + body: null, + reason: "A customer reply is not safe from the supplied context.", + }; + +const missingContext = missingContextFor(classification, normalized, request); +const matchedSignals = signalsFor(classification, normalized); +const hasDraftProposal = draftEmail.proposed !== false; +const sendGate = { + status: "requires_human_approval", + action: hasDraftProposal ? "send_customer_email" : "no_customer_send_proposed", + rationale: hasDraftProposal + ? "The draft is customer-ready, but this skill never sends. A separate governed send lane must approve delivery." + : "The request needs review or more evidence before a customer reply is safe.", +}; + +const result = { + classification, + severity, + confidence, + recommended_path: recommendedPath, + evidence: { + source: stringValue(request.source) ?? "inline_support_request", + source_summary: summarize(subject, body), + matched_signals: matchedSignals, + missing_context: missingContext, + taxonomy_coverage: ["how_to", "billing", "account_access", "bug", "abuse", "unknown"], + private_data_required: ["billing", "account_access"].includes(classification), + send_side_effects: "none", + }, + draft_email: draftEmail, + send_gate: sendGate, +}; + +process.stdout.write(`${JSON.stringify(result, null, 2)}\n`); + +function readInputs() { + if (process.env.RUNX_INPUTS_PATH) { + return JSON.parse(fs.readFileSync(process.env.RUNX_INPUTS_PATH, "utf8")); + } + if (process.env.RUNX_INPUTS_JSON) { + return JSON.parse(process.env.RUNX_INPUTS_JSON); + } + return { + support_request: parseInputValue(process.env.RUNX_INPUT_SUPPORT_REQUEST), + policy: parseInputValue(process.env.RUNX_INPUT_POLICY), + }; +} + +function parseInputValue(raw) { + if (raw === undefined || raw === "") return undefined; + try { + return JSON.parse(raw); + } catch { + return raw; + } +} + +function classify(text) { + if (matches(text, ["abuse", "spam", "phishing", "harassment", "threat", "fraud", "compromised"])) { + return "abuse"; + } + if (matches(text, ["invoice", "billing", "charge", "refund", "paid", "payment", "subscription", "plan", "tax"])) { + return "billing"; + } + if (matches(text, ["login", "password", "reset", "locked out", "2fa", "mfa", "owner", "access", "account"])) { + return "account_access"; + } + if (matches(text, ["error", "bug", "broken", "500", "failed", "crash", "exception", "does not work", "regression"])) { + return "bug"; + } + if (matches(text, ["how do i", "how can i", "where do i", "what should", "setup", "set up", "configure", "verify", "dns", "domain", "docs"])) { + return "how_to"; + } + return "unknown"; +} + +function classifySeverity(classification, text) { + if (classification === "abuse") return "high"; + if (classification === "billing" && matches(text, ["double charge", "charged twice", "refund"])) return "high"; + if (classification === "account_access") return "high"; + if (classification === "bug" && matches(text, ["all users", "production", "down", "data loss", "security"])) return "critical"; + if (classification === "bug") return "medium"; + if (classification === "unknown") return "medium"; + return "low"; +} + +function confidenceFor(classification, text) { + const signalCount = signalsFor(classification, text).length; + if (classification === "unknown") return 0.35; + if (signalCount >= 3) return 0.88; + if (signalCount === 2) return 0.78; + return 0.66; +} + +function recommendedPathFor(classification, confidence) { + if (confidence < 0.5) return "manual_review"; + switch (classification) { + case "how_to": + return "reply_draft"; + case "billing": + return "billing_review"; + case "account_access": + return "account_review"; + case "bug": + return "engineering_intake"; + case "abuse": + return "abuse_review"; + default: + return "request_info"; + } +} + +function buildDraftEmail({ request, subject, body, productName, signature, customerName }) { + const greeting = customerName ? `Hi ${customerName},` : "Hi,"; + const requestLine = summarize(subject, body); + const answer = answerForHowTo(`${subject ?? ""}\n${body ?? ""}`, productName); + return { + proposed: true, + subject: subject && /^re:/i.test(subject) ? subject : `Re: ${subject ?? "your support request"}`, + body: [ + greeting, + "", + `Thanks for the note. You asked about ${requestLine}.`, + "", + answer, + "", + "Before sending, an operator should confirm the product state and any account-specific details. This draft has not been sent.", + "", + "Thanks,", + signature, + ].join("\n"), + recipient_hint: stringValue(request.customer_email) ? "customer_email_present" : "no_customer_email", + }; +} + +function answerForHowTo(text, productName) { + const normalized = normalize(text); + if (matches(normalized, ["dns", "domain", "dkim", "spf", "dmarc", "verify"])) { + return `For ${productName} domain verification, check that the DNS records shown in the sending-domain setup are published exactly, then run the domain verification check again after DNS propagation. If a record still fails, compare the host/name and value fields character for character, including whether your DNS provider automatically appends the root domain.`; + } + if (matches(normalized, ["api", "webhook", "integration"])) { + return `For ${productName} integration setup, start by confirming the API key or webhook endpoint is scoped for the environment you are testing, then retry with one minimal request and save the response body if it fails.`; + } + return `For ${productName}, the safest next step is to follow the documented setup flow for the feature named in your request and confirm each required field before retrying. If the same step fails, send us the exact screen, error text, and timestamp so we can trace it.`; +} + +function missingContextFor(classification, text, request) { + const missing = []; + if (!stringValue(request.source)) missing.push("source locator"); + if (classification === "bug" && !matches(text, ["error", "500", "exception", "screenshot", "timestamp", "request id"])) { + missing.push("reproduction details or captured error"); + } + if (classification === "billing") missing.push("verified billing/account context"); + if (classification === "account_access") missing.push("verified account ownership context"); + if (classification === "unknown") missing.push("clear product surface and desired outcome"); + return missing; +} + +function signalsFor(classification, text) { + const dictionaries = { + how_to: ["how do i", "how can i", "where do i", "what should", "setup", "set up", "configure", "verify", "dns", "domain", "docs"], + billing: ["invoice", "billing", "charge", "refund", "paid", "payment", "subscription", "plan", "tax"], + account_access: ["login", "password", "reset", "locked out", "2fa", "mfa", "owner", "access", "account"], + bug: ["error", "bug", "broken", "500", "failed", "crash", "exception", "does not work", "regression"], + abuse: ["abuse", "spam", "phishing", "harassment", "threat", "fraud", "compromised"], + unknown: [], + }; + return (dictionaries[classification] ?? []).filter((signal) => text.includes(signal)); +} + +function summarize(subject, body) { + const candidate = subject || body || "the support request"; + const oneLine = String(candidate).replace(/\s+/g, " ").trim(); + return oneLine.length > 140 ? `${oneLine.slice(0, 137)}...` : oneLine; +} + +function matches(text, needles) { + return needles.some((needle) => text.includes(needle)); +} + +function normalize(value) { + return String(value ?? "").toLowerCase().replace(/\s+/g, " ").trim(); +} + +function firstName(value) { + if (!value) return null; + return value.split(/\s+/)[0]?.replace(/[^a-zA-Z'-]/g, "") || null; +} + +function stringValue(value) { + return typeof value === "string" && value.trim().length > 0 ? value.trim() : null; +} + +function objectValue(value, name) { + if (!value || typeof value !== "object" || Array.isArray(value)) { + fail(`${name} must be an object`); + } + return value; +} + +function fail(message) { + process.stderr.write(`${message}\n`); + process.exit(64); +} diff --git a/skills/write-harness/SKILL.md b/skills/write-harness/SKILL.md index dc2b748d1..97965091e 100644 --- a/skills/write-harness/SKILL.md +++ b/skills/write-harness/SKILL.md @@ -62,6 +62,11 @@ Start from the skill contract (SKILL.md + execution profile). Design fixtures fo invalid tool path. Expect failure with meaningful error. - **Governance gates** (composite skills only): one fixture per approval or policy transition that matters. +- **Publication evidence**: for skills intended for registry or public use, + include checks that prove the registry listing or public artifact is reachable, + durable, and tied to the submitted source. +- **User-value boundary**: include at least one assertion or acceptance check + that protects the real user-visible promise, not only internal step success. Each fixture tests one thing. Do not combine happy-path and error checks. Test the contract, not the internal wiring. @@ -83,6 +88,7 @@ internal builder transcript. That means: - explain catalog fit against adjacent current runx skills or graphs - describe the concrete user-visible artifact, not only the internal execution sequence +- name who would use, trust, link, install, or maintain the artifact and why - convert unresolved ambiguity into explicit maintainer decisions - keep issue comments, amendments, and approval records as provenance instead of copying them into the public proposal @@ -110,6 +116,9 @@ that in `maintainer_decisions` rather than leaking it into the fixture target. bundle. - Strategic bar: every fixture should protect a user-visible promise, trust boundary, or failure mode that matters for the skill's purpose. +- Public value bar: do not write fixtures for a skill whose only value is that it + exists. Return `not_first_party` or `needs_agent` when the proposal lacks a + credible user, operator, maintainer, catalog, or public proof value. - Stop conditions: return `needs_agent` when the contract is too vague to harness, and return `not_first_party` when the proposed skill should be reuse, Sourcey/content work, or a graph amendment instead. diff --git a/tests/github-thread.test.ts b/tests/github-thread.test.ts index 45738b84c..d14d813f3 100644 --- a/tests/github-thread.test.ts +++ b/tests/github-thread.test.ts @@ -528,6 +528,108 @@ describe("GitHub thread helper", () => { await rm(tempDir, { recursive: true, force: true }); } }); + + it("pushes Frantic open lifecycle operations through the GitHub adapter", async () => { + const tempDir = await mkdtemp(path.join(os.tmpdir(), "runx-frantic-github-")); + const ghBin = path.join(tempDir, "fake-gh.mjs"); + const logPath = path.join(tempDir, "gh.log"); + + try { + await writeFile(ghBin, fakeGhScript(logPath)); + await chmod(ghBin, 0o700); + const result = pushGitHubLifecycleIntent({ + thread: { + adapter: { + adapter_ref: "auscaster/frantic-board#issue/7", + }, + thread_locator: "github://auscaster/frantic-board/issues/7", + canonical_uri: "https://github.com/auscaster/frantic-board/issues/7", + metadata: { + repo: "auscaster/frantic-board", + }, + }, + outboxEntry: { + entry_id: "github:claim-1:thread.open", + kind: "provider_thread_lifecycle", + status: "pending", + metadata: { + action: "open", + }, + }, + env: { + ...process.env, + RUNX_GH_BIN: ghBin, + GH_FAKE_LOG: logPath, + GH_FAKE_ISSUE_STATE: "CLOSED", + }, + }); + + const calls = JSON.parse(await readFile(logPath, "utf8")); + expect(calls.map((call: { args: string[] }) => call.args.slice(0, 2).join(" "))).toEqual([ + "issue view", + "issue reopen", + ]); + expect(result).toMatchObject({ + outbox_entry: { + status: "published", + locator: "https://github.com/auscaster/frantic-board/issues/7", + }, + lifecycle: { + opened: true, + }, + }); + } finally { + await rm(tempDir, { recursive: true, force: true }); + } + }); + + it("publishes an already-open lifecycle operation without GitHub mutation", async () => { + const tempDir = await mkdtemp(path.join(os.tmpdir(), "runx-frantic-github-")); + const ghBin = path.join(tempDir, "fake-gh.mjs"); + const logPath = path.join(tempDir, "gh.log"); + + try { + await writeFile(ghBin, fakeGhScript(logPath)); + await chmod(ghBin, 0o700); + const result = pushGitHubLifecycleIntent({ + thread: { + adapter: { + adapter_ref: "auscaster/frantic-board#issue/7", + }, + thread_locator: "github://auscaster/frantic-board/issues/7", + canonical_uri: "https://github.com/auscaster/frantic-board/issues/7", + metadata: { + repo: "auscaster/frantic-board", + }, + }, + outboxEntry: { + entry_id: "github:claim-1:thread.open", + kind: "provider_thread_lifecycle", + status: "pending", + metadata: { + action: "open", + }, + }, + env: { + ...process.env, + RUNX_GH_BIN: ghBin, + GH_FAKE_LOG: logPath, + }, + }); + + const calls = JSON.parse(await readFile(logPath, "utf8")); + expect(calls.map((call: { args: string[] }) => call.args.slice(0, 2).join(" "))).toEqual([ + "issue view", + ]); + expect(result).toMatchObject({ + lifecycle: { + opened: true, + }, + }); + } finally { + await rm(tempDir, { recursive: true, force: true }); + } + }); }); function fakeGhScript(logPath: string): string { @@ -553,7 +655,7 @@ if (args[0] === "issue" && args[1] === "create") { } if (args[0] === "issue" && args[1] === "view") { process.stdout.write(JSON.stringify({ - state: "OPEN", + state: process.env.GH_FAKE_ISSUE_STATE || "OPEN", url: "https://github.com/auscaster/frantic-board/issues/7", labels: [{ name: "frantic:open" }] })); diff --git a/tests/init-command.test.ts b/tests/init-command.test.ts index 2b574c834..136c262fd 100644 --- a/tests/init-command.test.ts +++ b/tests/init-command.test.ts @@ -7,7 +7,7 @@ import { describe, expect, it } from "vitest"; import { runCli } from "../packages/cli/src/index.js"; describe("runx init", () => { - it("scaffolds a new authoring package through runx new", async () => { + it("scaffolds a native skill through runx new", async () => { const tempDir = await mkdtemp(path.join(os.tmpdir(), "runx-new-package-")); const stdout = createMemoryStream(); const stderr = createMemoryStream(); @@ -25,31 +25,23 @@ describe("runx init", () => { readonly new: { readonly action: string; readonly name: string; - readonly packet_namespace: string; readonly directory: string; readonly files: readonly string[]; }; }; const target = path.join(tempDir, "docs-demo"); expect(report.new).toMatchObject({ - action: "package", + action: "skill", name: "docs-demo", - packet_namespace: "docs.demo", directory: target, }); expect(report.new.files).toContain("SKILL.md"); + expect(report.new.files).toContain("X.yaml"); + expect(report.new.files).toContain("run.mjs"); await expect(readFile(path.join(target, "SKILL.md"), "utf8")).resolves.toContain("name: docs-demo"); - await expect(readFile(path.join(target, "X.yaml"), "utf8")).resolves.toContain("tool: docs.echo"); - await expect(readFile(path.join(target, "tools/docs/echo/fixtures/basic.yaml"), "utf8")).resolves.toContain("lane: deterministic"); - await expect(readFile(path.join(target, "fixtures/agent.yaml"), "utf8")).resolves.toContain("lane: agent"); - await expect(readFile(path.join(target, "fixtures/agent.replay.json"), "utf8")).resolves.toContain("runx.replay.v1"); - await expect(readFile(path.join(target, "dist/packets/echo.v1.schema.json"), "utf8")).resolves.toContain("docs.demo.echo.v1"); - const manifest = JSON.parse(await readFile(path.join(target, "tools/docs/echo/manifest.json"), "utf8")) as { - readonly source_hash?: string; - readonly schema_hash?: string; - }; - expect(manifest.source_hash).toMatch(/^sha256:[a-f0-9]{64}$/); - expect(manifest.schema_hash).toMatch(/^sha256:[a-f0-9]{64}$/); + await expect(readFile(path.join(target, "X.yaml"), "utf8")).resolves.toContain("type: cli-tool"); + await expect(readFile(path.join(target, "run.mjs"), "utf8")).resolves.toContain("RUNX_INPUT_MESSAGE"); + await expect(stat(path.join(target, "package.json"))).rejects.toMatchObject({ code: "ENOENT" }); } finally { await rm(tempDir, { recursive: true, force: true }); } diff --git a/tests/official-skill-catalog.test.ts b/tests/official-skill-catalog.test.ts index e06da8798..05c9e288a 100644 --- a/tests/official-skill-catalog.test.ts +++ b/tests/official-skill-catalog.test.ts @@ -17,6 +17,7 @@ import { resolveRunxBinary } from "./runx-binary.js"; const publicCatalogPackages = [ "brand-voice", + "business-ops", "charge", "content-pipeline", "deep-research-brief", diff --git a/tests/registry-fixtures.ts b/tests/registry-fixtures.ts index 1281bd69b..2bc7d62b5 100644 --- a/tests/registry-fixtures.ts +++ b/tests/registry-fixtures.ts @@ -38,7 +38,7 @@ export interface RegistrySkillVersion { readonly trust_tier: RegistryTrustTier; readonly maturity?: "alpha" | "beta" | "stable"; readonly catalog_kind?: "skill" | "graph"; - readonly catalog_audience?: "public" | "builder" | "operator"; + readonly catalog_audience?: "public" | "builder" | "operator" | "system"; readonly catalog_visibility?: "public" | "internal"; readonly attestations?: readonly RegistryAttestation[]; readonly required_scopes: readonly string[]; diff --git a/tests/scafld-skill-parser.test.ts b/tests/scafld-skill-parser.test.ts index d2cbe0dc2..e1c0d3f8f 100644 --- a/tests/scafld-skill-parser.test.ts +++ b/tests/scafld-skill-parser.test.ts @@ -36,6 +36,8 @@ describe("scafld graph stage contract", () => { expect(wrapper).toContain('"build"'); expect(wrapper).toContain('"build_to_review"'); expect(wrapper).toContain('"handoff"'); + expect(wrapper).toContain("scafld_min_version"); + expect(wrapper).toContain("ensureScafldVersion"); expect(wrapper).toContain("function runBuildToReview"); expect(wrapper).not.toContain('"new"'); expect(wrapper).not.toContain('"branch"'); @@ -72,6 +74,10 @@ describe("scafld graph stage contract", () => { `#!/usr/bin/env node const argv = process.argv.slice(2); const command = argv[0] || ""; +if (command === "--version") { + process.stdout.write("2.4.0\\n"); + process.exit(0); +} if (command === "review") { process.stderr.write("scafld review[command] started node reviewer.mjs\\n"); process.stderr.write("scafld review[command] completed exit=0 elapsed=4ms last_output=0s\\n"); @@ -130,4 +136,52 @@ process.exit(1); await rm(tempDir, { recursive: true, force: true }); } }); + + it("fails closed when the resolved scafld is older than 2.4.0", async () => { + const tempDir = await mkdtemp(path.join(os.tmpdir(), "runx-scafld-old-version-")); + const fakeScafld = path.join(tempDir, "fake-scafld.mjs"); + const wrapperPath = path.join(scafldStageDir, "run.mjs"); + + try { + await writeFile( + fakeScafld, + `#!/usr/bin/env node +const argv = process.argv.slice(2); +const command = argv[0] || ""; +if (command === "--version") { + process.stdout.write("2.3.12\\n"); + process.exit(0); +} +if (command === "validate") { + process.stdout.write(JSON.stringify({ + ok: true, + command: "validate", + result: { task_id: argv[1], valid: true, errors: null }, + }) + "\\n"); + process.exit(0); +} +process.stderr.write(\`unsupported command: \${command}\\n\`); +process.exit(1); +`, + { mode: 0o755 }, + ); + + await expect(execFile("node", [wrapperPath], { + cwd: tempDir, + env: { + ...process.env, + RUNX_INPUTS_JSON: JSON.stringify({ + command: "validate", + task_id: "fixture-task", + fixture: tempDir, + scafld_bin: fakeScafld, + }), + }, + })).rejects.toMatchObject({ + stderr: expect.stringContaining("scafld 2.4.0 or newer is required"), + }); + } finally { + await rm(tempDir, { recursive: true, force: true }); + } + }); }); diff --git a/tools/thread/frantic_thread_outbox.d.mts b/tools/thread/frantic_thread_outbox.d.mts index 3fb799cdd..160334854 100644 --- a/tools/thread/frantic_thread_outbox.d.mts +++ b/tools/thread/frantic_thread_outbox.d.mts @@ -1,5 +1,5 @@ export interface FranticThreadIntent { - readonly kind: "thread.create" | "thread.comment" | "thread.labels" | "thread.close"; + readonly kind: "thread.create" | "thread.comment" | "thread.labels" | "thread.close" | "thread.open"; readonly outbox_id: string; readonly provider: string; readonly thread_locator?: string; diff --git a/tools/thread/frantic_thread_outbox.mjs b/tools/thread/frantic_thread_outbox.mjs index e5f0d17c8..ebaec8b8f 100644 --- a/tools/thread/frantic_thread_outbox.mjs +++ b/tools/thread/frantic_thread_outbox.mjs @@ -70,7 +70,7 @@ export function normalizeFranticThreadIntent(intent) { const bountyNumber = requiredPositiveInteger(intent.bounty_number, "intent.bounty_number"); const occurredAt = requiredString(intent.occurred_at, "intent.occurred_at"); - if (!["thread.create", "thread.comment", "thread.labels", "thread.close"].includes(kind)) { + if (!["thread.create", "thread.comment", "thread.labels", "thread.close", "thread.open"].includes(kind)) { throw new Error(`unsupported Frantic thread intent kind '${kind}'.`); } @@ -240,7 +240,7 @@ function buildGitHubOutboxEntry(intent) { channel: "github_issue", source: "frantic", source_ref: intent.source_ref, - action: intent.kind === "thread.labels" ? "labels" : "close", + action: intent.kind === "thread.labels" ? "labels" : intent.kind === "thread.open" ? "open" : "close", add_labels: intent.add_labels, remove_labels: intent.remove_labels, close_reason: intent.reason, diff --git a/tools/thread/github_adapter.mjs b/tools/thread/github_adapter.mjs index 38baf3e9d..e565045ba 100644 --- a/tools/thread/github_adapter.mjs +++ b/tools/thread/github_adapter.mjs @@ -916,13 +916,19 @@ export function pushGitHubLifecycleIntent({ }); const addLabels = stringList(metadata.add_labels); const removeLabels = stringList(metadata.remove_labels); + const lifecycleAction = firstNonEmptyString(metadata.action); const closeReason = firstNonEmptyString(metadata.close_reason, "completed"); if (!repoSlug) { throw new Error("GitHub issue repo slug is required to push a provider thread lifecycle entry."); } - if (addLabels.length === 0 && removeLabels.length === 0 && metadata.action !== "close") { - throw new Error("provider_thread_lifecycle requires label changes or a close action."); + if ( + addLabels.length === 0 && + removeLabels.length === 0 && + lifecycleAction !== "close" && + lifecycleAction !== "open" + ) { + throw new Error("provider_thread_lifecycle requires label changes, an open action, or a close action."); } const labelResult = applyGitHubLabelChanges({ @@ -935,7 +941,8 @@ export function pushGitHubLifecycleIntent({ env, }); - const shouldClose = firstNonEmptyString(metadata.action) === "close"; + const shouldClose = lifecycleAction === "close"; + const shouldOpen = lifecycleAction === "open"; if (shouldClose && String(issueState.state ?? "").toUpperCase() !== "CLOSED") { closeGitHubIssue({ repoSlug, @@ -945,6 +952,14 @@ export function pushGitHubLifecycleIntent({ env, }); } + if (shouldOpen && String(issueState.state ?? "").toUpperCase() === "CLOSED") { + reopenGitHubIssue({ + repoSlug, + issueNumber: issueRef.issue_number, + cwd, + env, + }); + } return { outbox_entry: prune({ @@ -967,6 +982,7 @@ export function pushGitHubLifecycleIntent({ added_labels: labelResult.addedLabels, removed_labels: labelResult.removedLabels, closed: shouldClose, + opened: shouldOpen, close_reason: shouldClose ? closeReason : undefined, }), }; @@ -1261,6 +1277,31 @@ function closeGitHubIssue({ repoSlug, issueNumber, reason, cwd, env }) { }, { tokenFallback: true }); } +function reopenGitHubIssue({ repoSlug, issueNumber, cwd, env }) { + if (canUseGitHubRest(env)) { + runGitHubRest({ + method: "PATCH", + path: gitHubIssueApiPath(repoSlug, issueNumber), + env, + body: { + state: "open", + }, + acceptedStatuses: [200], + }); + return; + } + runGhCommand([ + "issue", + "reopen", + issueNumber, + "--repo", + repoSlug, + ], { + cwd, + env, + }, { tokenFallback: true }); +} + function selectExistingGitHubMessageOutboxEntry(thread, outboxEntry) { const existingOutbox = Array.isArray(thread.outbox) ? thread.outbox.filter(isRecord) : []; const requestedMetadata = optionalRecord(outboxEntry.metadata) ?? {};