agentleFS
khangledoanminhkhangledoanminh/agentskillslab/skills/documentation-engineer/SKILL.md

documentation-engineer

Skill1 starsChanged 3 months agoMITView on GitHub

Assess and repair documentation quality with measurable checks: missing docstrings, undocumented public APIs, dead links, stale examples, and readability scoring, then generate documentation that matches the current code exactly. Use when docs are missing, outdated, or an onboarding guide is needed.

6 other projects use a version of this.

Adopted by

Most recently taken up by

The file

Files
SKILL.md

Documentation Engineer

Docs phải khớp code THẬT tại thời điểm viết. Pipeline: audit coverage thật → tìm lỗ hổng → viết từ source code hiện tại → verify links + examples chạy được.

Quy trình (5 bước)

  1. Audit coverage: chạy scripts/doc_coverage.py <repo> — % public functions/classes có docstring, % modules có README, dead links trong markdown.
  2. Prioritize gaps: public API không có doc > nội bộ > ví dụ cũ. Ưu tiên API public trước.
  3. Generate docs từ code thật: parse signature + docstring hiện có + source body; viết docs mô tả đúng behavior hiện tại. KHÔNG extrapolate tính năng chưa có trong code.
  4. Verify examples: mỗi code example trong docs phải chạy được (script chạy example, kiểm exit code).
  5. Verify links: scan mọi link trong docs → target tồn tại (file local phải tồn tại; URL external đánh dấu cần manual verify).

Metrics mặc định

MetricTarget
Public API doc coverage≥ 90%
Dead internal links0
Runnable examples100% ví dụ có code block

Nguyên tắc

  • Docs mô tả behavior hiện tại; nếu code sai, báo bug riêng — không viết docs theo ý muốn.
  • Ví dụ phải chạy; ví dụ không chạy worse hơn không có ví dụ.
  • Markdown nhất quán: heading levels, code fence language.

Scripts

  • scripts/doc_coverage.py <repo> [--output report.json]
  • scripts/check_examples.py <docs-dir> — chạy code blocks trong markdown

References

  • STYLE-GUIDE.md
  • EXAMPLE-RULES.md

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.