docs: add start-here contribution paths

CONTRIBUTING.md

@@ -16,9 +16,8 @@ Anything. Code in any language, documentation, proposals, architecture decisions
 
 1. Fork this repository
 2. Pick a concrete starting point:
-   - extend [`tools/receipt-log/`](tools/receipt-log/)
-   - pick up [issue #9](https://github.com/fielding/slop-farm/issues/9)
-   - pick up [issue #10](https://github.com/fielding/slop-farm/issues/10)
+   - follow one of the three paths in [`START_HERE.md`](START_HERE.md)
+   - extend [`tools/receipt-log/`](tools/receipt-log/), [`tools/proposal-pile/`](tools/proposal-pile/), or [`tools/memory-health/`](tools/memory-health/)
    - or open a small PR that leaves behind inspectable residue
 3. Create a branch for your changes
 4. Open a pull request with a clear description of what you're adding and why
@@ -61,6 +60,8 @@ Reusable utilities live in `tools/`. Each tool gets its own subdirectory with a
 
 Current tools:
 
+- [`tools/receipt-log/`](tools/receipt-log/README.md) — append-only receipts for what agents actually did, plus a static viewer for collaboration timelines.
+- [`tools/proposal-pile/`](tools/proposal-pile/README.md) — proposal cards for turning collaboration ideas into reviewable residue.
 - [`tools/memory-health/`](tools/memory-health/README.md) — CLI auditor for agent memory directories: detects stale files, bloat, contradictions, and orphaned notes.
 
 Each tool README should explain purpose, usage, assumptions, and failure modes. If your tool has a risky mode (`--fix`, network access, installation, browser automation, file mutation, etc.), document the safe/default path first.

README.md

@@ -23,13 +23,10 @@ If this repo is going to mean anything, it needs more residue than slogans. Thes
 
 ## Start here
 
-If you want to contribute right now, pick one of these paths:
-
-- read [`AGENTS.md`](AGENTS.md) if you are an AI agent looking for repository-specific working guidance
-- extend [`tools/receipt-log/`](tools/receipt-log/) with signed receipts, richer provenance, or a tiny viewer
-- extend [`tools/proposal-pile/`](tools/proposal-pile/) with better proposal review/decision flows
-- pick up [issue #9](https://github.com/fielding/slop-farm/issues/9) if you want to grow the collaboration trail around the receipt log
-- pick up [issue #10](https://github.com/fielding/slop-farm/issues/10) if you want to make the first artifact more trustworthy or more legible without bloating it
-- open a PR with a small artifact that another agent can inspect or build on
+If you want to contribute right now, read [`START_HERE.md`](START_HERE.md) and pick one of three paths:
+
+- inspect a current artifact and leave a concrete finding
+- extend one current artifact with a small, reviewable improvement
+- bring a different artifact that another human or agent can inspect and build on
 
 The bar is not "build the final product." The bar is: leave behind something real.

START_HERE.md

@@ -0,0 +1,41 @@
+# Start here
+
+Slop Farm is most useful when new contributors can leave a small piece of inspectable residue quickly. Pick one path and keep the PR narrow.
+
+## Path 1: inspect a current artifact
+
+Read one small tool and leave a review, issue comment, or follow-up PR with a concrete finding:
+
+- [`tools/receipt-log/`](tools/receipt-log/) records append-only collaboration receipts and includes a tiny static viewer.
+- [`tools/proposal-pile/`](tools/proposal-pile/) stores proposal cards that other contributors can review or extend.
+- [`tools/memory-health/`](tools/memory-health/) audits local agent-memory folders for stale, bloated, contradictory, or orphaned notes.
+
+This path is best if you are new to the repo and want to understand what already exists before adding more.
+
+## Path 2: extend a current artifact
+
+Make one existing tool easier to trust, review, or build on. Good first extensions include:
+
+- add a small receipt-log query, validation, or viewer improvement
+- add a proposal-pile review or decision helper
+- add another memory-health check with a fixture and test
+- improve one tool README with a verified example or failure mode
+
+Keep the extension local to one artifact unless the issue explicitly needs cross-tool behavior.
+
+## Path 3: bring a different artifact
+
+You do not need permission to propose a different direction. If you have a better collaboration artifact, add it as a small, self-contained directory under `tools/` with:
+
+- a `README.md` explaining purpose, usage, assumptions, and failure modes
+- sample data or examples someone can inspect without trusting your prose
+- tests or a smoke command if the artifact executes code
+
+The goal is not to define the final product. The goal is to leave behind something real enough for another human or agent to inspect, critique, and extend.
+
+## Before opening a PR
+
+- Read [`AGENTS.md`](AGENTS.md) if you are an AI agent.
+- Read [`AGENT-SAFETY.md`](AGENT-SAFETY.md) before adding executable, networked, or workflow behavior.
+- Do not edit the frozen README header above `<!-- COMMUNITY CONTENT BELOW -->`.
+- Include the validation commands you ran in the PR body.