project-documentation-archivist

Distills durable project knowledge from documentation and GitHub sources into canonical Markdown files.

Updated Jun 11, 2026
One-click install
npx skills add https://github.com/coferlandia/coferlandia-skills --skill project-documentation-archivist-coferlandia
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: project-documentation-archivist
Source: https://github.com/coferlandia/coferlandia-skills/tree/main/skills/content/project-documentation-archivist
Command: npx skills add https://github.com/coferlandia/coferlandia-skills --skill project-documentation-archivist-coferlandia

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve? Project knowledge gets scattered across README files, legacy TODO/HISTORY documents, GitHub Issues, PRs, and commits, leaving agents and developers without a reliable source of durable truth. This Skill organizes that sprawl into maintained canonical files (README, AGENTS, DECISIONS, RUNBOOK) while migrating legacy operational tracking into GitHub Issues. ## Core Features & Use Cases - Knowledge distillation: Reads local documentation and GitHub sources (Issues, PRs, commits) and routes only durable knowledge into README.md, AGENTS.md, DECISIONS.md, and RUNBOOK.md. - Legacy migration: Migrates TODO.md, HISTORY.md, and OPEN_QUESTIONS.md into GitHub Issues through a deterministic, dry-run-by-default script with semantic classification and cutover validation. - Traceability catalog: Maintains .agent/catalog/SOURCE_INDEX.md and PROCESSING_RUNS.md so every processed source and run is auditable. - Use Case: A repository still tracks work in TODO.md and HISTORY.md. Run the migration workflow to classify each legacy entry, create or link GitHub Issues, distill knowledge-only content into canonical docs, and validate the cutover before deleting the legacy files. ## Quick Start Ask the agent to inventory this project's documentation and GitHub sources, then update README, AGENTS, DECISIONS, and RUNBOOK with the durable knowledge it finds.

Frequently Asked Questions about project-documentation-archivist

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I migrate TODO.md and HISTORY.md to GitHub Issues?▼

Run github_migration.py preflight and inventory, classify each item with a disposition such as CREATE_OPEN_ISSUE or KNOWLEDGE_ONLY, validate the decisions file, then run apply with --apply after authorization. Finish with validate-cutover before deleting the legacy files.

How do I keep AGENTS.md and README.md up to date from GitHub activity?▼

Run the normal knowledge-distillation workflow: inventory local and GitHub sources, read them deeply, and route only durable knowledge into README, AGENTS, DECISIONS, and RUNBOOK. Operational work state stays in GitHub Issues and is never mirrored into Markdown.

Does the migration script create duplicate GitHub Issues on reruns?▼

No. Created Issues carry an invisible migration marker, and the script persists a local journal immediately after each creation. Reruns prefer the journal and fall back to marker search, and marker lookup fails closed on GitHub read errors.

What are the requirements for running this documentation workflow?▼

You need read/write access to the target project, Python 3.11 or later, Git when the project is versioned, and an authenticated GitHub CLI (gh) for GitHub ingestion or migration operations.

When should content go to DECISIONS.md instead of a GitHub Issue?▼

Durable rationale, trade-offs, and confirmed present-state facts belong in DECISIONS.md or other canonical files. Only actionable future work or material unresolved questions should become GitHub Issues; knowledge-only content must not manufacture an Issue.