documentation-writer

SkillDocs & knowledge

Creates structured technical documentation from code, requirements, or existing notes. Use when a feature, module, or system needs written docs and none exist yet, or existing ones are out of date.

Then ask your AI: use the documentation-writer skill

Details

Instructions available. Your AI can read the instructions. Execution depends on the setup they require.

Add it as a step in a pipeline you build on Ahel, or connect Ahel to Claude, ChatGPT, Cursor, Claude Code or Codex and ask your AI to use it.

What this skill tells your AI

The instructions your AI receives, as published by codebygarv/ai-skills in skills/documentation/documentation-writer/SKILL.md and read by Ahel’s review.

Purpose

Produce clear, structured technical documentation from a given source — source code, requirements, meeting notes, or a mix — that explains what something is, how it works, and how to use it, for the audience it's meant for.

When to Use

  • A feature, module, API, or system has no documentation, or what exists is stale.
  • Turning requirements or design notes into a reference doc.
  • The user has working code but needs it explained for other engineers, or non-engineers.

What to Analyze / Do

  1. Identify the audience — engineers integrating with this, engineers maintaining it, or non-technical stakeholders — and match depth/jargon accordingly. Ask if unclear and it meaningfully changes the doc.
  2. Establish structure before writing prose — overview, how it works, usage/examples, configuration, edge cases/limitations, is a reasonable default shape; adapt to what the source material actually supports.
  3. Extract behavior from source — if working from code, document what it actually does, not an idealized version of what it's supposed to do; flag any place the code and stated intent seem to diverge.
  4. Include concrete examples — a real usage example beats an abstract description every time.
  5. Note what's deliberately left out — assumptions, out-of-scope behavior, known limitations — so the doc doesn't imply completeness it doesn't have.

Output Format

  • A structured Markdown doc: title, one-paragraph overview, then sections appropriate to the content (How it works / Usage / Configuration / Examples / Limitations).
  • Code examples in fenced blocks with the correct language tag.
  • Headings that let someone scan and jump to the relevant section rather than reading linearly.

Avoid

  • Documenting intended behavior instead of actual behavior when working from code — verify against the code, don't assume the docstring/comment is accurate.
  • Padding with generic boilerplate paragraphs that don't convey real information.
  • Writing at the wrong depth for the stated audience (over-explaining basics to engineers, or under-explaining to non-technical readers).

Signals

GitHub stars
27
Forks
1
Last commit
Aug 2026
Advanced
Item type
skill
Key
documentation-writer-codebygarv
Source
github.com/codebygarv/ai-skills