tasks.md

Lash Development Tasks - Master Index

This directory contains the comprehensive task breakdown for implementing Lash v1.0.

Project Status

Current Phase: v1.0 COMPLETE Target: v1.0 Production Release Estimated Duration: 40-60 days (8-13 weeks) Completed: Project Setup, Core Data Model, Error Handling, Markdown Parser, Linter & Formatter, SQLite Schema, Indexing, Dependency Resolution, CLI Framework, CLI Commands, Fuzzy Search, Agent Integration, TUI, Testing, Documentation

Task Organization

Tasks are organized by major module/component. Each task file contains:

• Task descriptions with priority levels
• Dependencies on other tasks
• Estimated effort
• Success criteria
• Subtasks broken down to actionable units

Task Files

Foundation & Infrastructure

• tasks.project-setup.md - Project initialization, tooling, error taxonomy ✅
• tasks.core-data-model.md - Core data structures (Task, File, Dependency, Label models) ✅
• tasks.error-handling.md - Error types, formatting, aggregation (Tasks 1-3 complete) ✅

Parsing & Validation

• tasks.markdown-parser.md - Markdown parsing implementation ✅
• tasks.linter.md - Linting rules, validation, auto-formatting ✅
• tasks.doc-annotation.md - @doc annotation for semantic documentation links ✅

Database Layer

• tasks.sqlite-schema.md - SQLite schema, repositories, queries ✅
• tasks.indexing.md - File scanning, indexing engine, verification ✅
• tasks.dependency-resolution.md - Dependency graph, cycle detection, state computation, incremental updates ✅

CLI & Commands

• tasks.cli-framework.md - CLI infrastructure, output formatting, progress reporting (Tasks 1-8 complete) ✅
• tasks.cli-commands.md - Individual command implementations: lint, format, index, check-index, list, show, search, graph, check-links, agent-prompt (Tasks 1-14 complete) ✅
• tasks.fuzzy-search.md - Full-text search, ranking, search command (Tasks 1-5 complete) ✅
• tasks.cli-task-mutation.md - Task mutation & display gaps: add --id/--depends-on validation, waive, show full record, update (GitHub #23–#27) ✅

User Interfaces

• tasks.tui.md - Terminal UI implementation ✅
• tasks.task-creation-ui.md - Task creation via CLI (lash add) and TUI form modal (Tasks 1-12 complete) ✅
• tasks.status-bar-activity.md - Status bar in-progress + recently-completed activity sections ✅
• tasks.live-updates.md - Live TUI updates (Store actor, file watcher, external-change reload) ✅

Advanced Features

• tasks.agent-integration.md - Agent prompts, token minimization (Tasks 1-6 complete) ✅
• tasks.agent-skill-install.md - lash skill install for Claude/Codex/Cursor/AGENTS.md targets ✅

Content Features

• tasks.description-section.md - ## Description section support (Tasks 1-8 complete) ✅
• tasks.contextual-notes.md - Contextual notes (plain bullets as inline requirements/criteria) ✅

Quality & Documentation

• tasks.testing.md - Test strategy, integration tests, benchmarks, CI/CD (Tasks 1-9 complete) ✅
• tasks.documentation.md - User guide, developer docs, examples, README ✅

Implementation Phases

Phase 1: Foundation (Weeks 1-2) - COMPLETE ✅

Goal: Set up project infrastructure and core data model

1. ✅ tasks.project-setup (ALL) - Complete
2. ✅ tasks.core-data-model (ALL) - Complete
3. ✅ tasks.error-handling#1 - Complete

Deliverable: Compiling project with core types; can create tasks programmatically Status: Complete. Project infrastructure established with 6 crates, error taxonomy, and core data model.

Phase 2: Core Functionality (Weeks 3-5) - COMPLETE ✅

Goal: Parse Markdown files and implement linting/validation

1. ✅ tasks.markdown-parser (ALL) - Complete
2. ✅ tasks.linter (ALL) - Complete

Deliverable: Can parse task files into data structures; lint and format files Status: Parser complete (390+ tests, 67.7µs benchmark). Linter complete with 20 rules, auto-formatter, and CLI integration (607 total tests).

Phase 3: Database & Indexing (Weeks 6-7) - COMPLETE ✅

Goal: Build database schema and indexing system

1. ✅ tasks.sqlite-schema#1-2 (define schema, create tables) - Complete
2. ✅ tasks.sqlite-schema#3-6 (repositories, queries) - Complete
3. ✅ tasks.indexing#0-6 (project root, file scanning, diff, indexing engine, verification, dependency updates, performance optimization) - Complete

Deliverable: lash index command works; can query database Status: Complete. Indexing engine production-ready with 238 tests passing. Performance exceeds all targets by 8-12x (10.5ms for 10 files, 61ms for 100 files, 425ms for 1000 files).

Phase 4: Dependencies & Queries (Weeks 7-8) - COMPLETE ✅

Goal: Resolve dependencies; implement query commands

1. ✅ tasks.dependency-resolution#1-7 (graph, cycles, resolution, status, blockers, export, incremental updates) - Complete
2. ✅ tasks.cli-framework#1-8 (CLI infrastructure: arg parsing, root detection, output formatting, progress, config, logging, command execution, exit codes) - Complete
3. ✅ tasks.cli-commands#1-6 (lint, format, index, check-index, list, show commands) - Complete
4. tasks.linter#4 (dependency validation) - Deferred to Phase 5

Deliverable: Can query tasks; dependency resolution works Status: Complete. All core CLI commands implemented with 862 tests passing. Dependency resolution (495 tests), CLI framework (626 tests), and command implementations (1,056 total tests across all crates) all working.

Phase 5: Search & Advanced Commands (Weeks 9-10) - COMPLETE ✅

Goal: Fuzzy search; additional features

1. ✅ tasks.fuzzy-search (ALL) - Complete
2. ✅ tasks.cli-commands#7-14 (search, graph, check-links, agent-prompt, advanced features) - Complete
3. ✅ tasks.dependency-resolution#4-5 - Complete

Deliverable: All core CLI commands functional Status: Complete. All advanced commands implemented including search (FTS5), graph (ASCII/DOT/Mermaid/JSON), check-links with fuzzy fix suggestions. 1,604+ total tests passing.

Phase 6: Agent Integration (Week 11) - COMPLETE ✅

Goal: Make Lash usable by AI agents

1. ✅ tasks.agent-integration#1-6 (schema, prompts, token utils, sparse context, CLI command, workflows) - Complete
2. tasks.error-handling#2-4 - Deferred

Deliverable: Agents can use Lash effectively Status: Complete. Agent integration production-ready with schema generation, prompt templates, token minimization, sparse context generation, lash agent-prompt command, and comprehensive workflow documentation. 40 tests + 22 doctests passing.

Phase 7: TUI (Week 12) - COMPLETE ✅

Goal: Terminal UI for interactive use

1. ✅ tasks.tui (ALL) - Complete (Task 5 Agent View deferred to future)

Deliverable: lash tui command provides interactive interface Status: Complete. Full TUI with navigation pane, detail view, search modal, label filter, theme selector (300+ Gogh schemes), tree view with expand/collapse. Task 5 (Agent View Mode) deferred intentionally.

Phase 8: Testing & Documentation (Week 13) - COMPLETE ✅

Goal: Polish and prepare for release

1. ✅ tasks.testing (ALL) - Complete
2. ✅ tasks.documentation#1-8 - Complete

Deliverable: Production-ready v1.0 release Status: Complete. Testing: 1,604+ tests, >80% coverage, CI/CD configured. Documentation: README, user guide, developer guide, agent guide, error code reference, examples, and Rustdoc coverage enforced with #![warn(missing_docs)] across all crates.

Phase 9: Content Features (Week 14) - COMPLETE ✅

Goal: Add description section feature for richer file context

1. ✅ tasks.description-section#1-6 (design doc, parser, linter, schema, indexing, FTS5 search) - Complete
2. ✅ tasks.description-section#7 (CLI commands: show, list --show-descriptions, agent-prompt --include-descriptions) - Complete
3. ✅ tasks.description-section#8 (TUI display) - Complete

Deliverable: Task files support optional ## Description section with length validation and full-text search Status: Complete. All description section features implemented and tested.

Critical Path

The following tasks form the critical path and must be completed in order:

Foundation → Parsing → Linting → Schema → Indexing → Dependencies → Commands → Polish

Parallelization opportunities exist within each phase (see individual task files for details).

Success Criteria for v1.0 Release

Must Have

• ✅ Parse valid Lash Markdown files
• ✅ Lint files with clear error messages
• ✅ SQLite database schema with repositories
• ✅ Index files into SQLite database
• ✅ Full dependency resolution and cycle detection
• ✅ lash list, lash show commands work
• ✅ lash search command with full-text search and filters
• ✅ lash graph exports dependency graph (ASCII, DOT, Mermaid, JSON formats)
• ✅ Agent integration (lash agent-prompt)
• ✅ Comprehensive test coverage (>80%)
• ✅ User documentation

Should Have

• ✅ TUI for interactive use (Tasks 1-4, 6-10 complete; Task 5 Agent View deferred)
• ✅ Incremental indexing
• ✅ Auto-formatting
• ✅ Performance benchmarks

Nice to Have

• ⚠️ Advanced agent views (deferred to future)
• ✅ Fuzzy link fixing (check-links --fix with suggestions)
• ⚠️ Archive command
• ✅ ## Description section for task files (all features complete)

Supporting Documentation

The following analysis documents informed this task breakdown:

• /docs/design-doc.md - Comprehensive design specification
• /docs/dependency-graph-architecture.md - Graph architecture design
• /docs/indexing-architecture.md - Indexing engine architecture

Notes

• All time estimates are rough guidelines; adjust based on actual progress
• Dependencies between tasks are clearly marked in individual task files
• Priority levels: CRITICAL (must have), HIGH (should have), MEDIUM (nice to have), LOW (future)
• Task files use checkbox format to track progress as work proceeds