Skip to content
AI & Agents
Skill

/documentation-writer

Write or update technical documentation for code, APIs, or systems. Use when asked to document a module, write a README, generate API docs, or update existing documentation.

BOOST
From plugin
atlas-agents
58713 skills
Install
$ npx -y skills add agulli/atlas-agents --skill documentation-writer --agent claude-code

How it fires

How this skill gets triggered: by you, by Claude, or both.

  • Fires itselfAuto-invocation. Claude auto-loads it when your prompt matches the work.Auto-invocation is when the right skill fires by itself at the right moment, driven by a FLOW.md router and a hook, instead of you invoking it by name. It is the difference between a skill being installed and a skill actually getting used.Read the full definition →
  • You can call itInvoke it directly when you want it.
  • Slash command/documentation-writer

Context preview

The summary Claude sees to decide when to auto-load this skill.

Write or update technical documentation for code, APIs, or systems. Use when asked to document a module, write a README, generate API docs, or update existing documentation.

SKILL.md

documentation-writer.SKILL.md
name: documentation-writer
description: Write or update technical documentation for code, APIs, or systems. Use when asked to document a module, write a README, generate API docs, or update existing documentation.
license: MIT

Overview

You write documentation that developers will actually read. That means concrete examples over abstract descriptions, and answering the question "how do I use this?" before "what is this?".

Process

1. **Read the source first.** Load the actual code before writing a single word. Documentation that doesn't match the implementation is worse than no documentation.

2. **Identify the documentation type:**

  • **README** — for projects and repositories
  • **Module/API docs** — for functions, classes, endpoints
  • **Architecture docs** — for system design and component relationships
  • **Runbook** — for operational procedures

3. **For READMEs**, follow this structure:

   # Project Name
   One sentence: what it does and who it's for.

   ## Quick Start
   The fewest possible steps to get something working.
   Code first. Explanation second.

   ## Installation
   ## Usage
   ## Configuration
   ## API Reference (if applicable)
   ## Contributing
   ## License

4. **For function/class docs**, write docstrings that include:

  • What the function does (one line)
  • Parameters: name, type, description, whether optional
  • Return value: type and description
  • Exceptions it may raise
  • One usage example

5. **Write the example first.** If you cannot write a concrete 5-line example of how to use something, you don't understand it well enough to document it. Go back to step 1.

6. **Verify accuracy.** Read your documentation and check every claim against the code. If you wrote "`returns a list of strings`" and the code returns a dict, fix it.

Rationalizations

| Excuse | Rebuttal | |--------|----------| | "The code is self-documenting" | No code is self-documenting to someone seeing it for the first time. Write the docs. | | "I'll add examples later" | Documentation without examples is theory. Write the example now. | | "The function signature explains it" | Type hints say what the types are, not what the function does or why. |

Verification

  • [ ] Every documented function has at least one usage example
  • [ ] Parameter types and return types are correct (verified against the source)
  • [ ] Quick Start section in README can be followed by a new developer with no prior context
  • [ ] No documentation contradicts the actual implementation
Read more
Ships withatlas-agents

This repository contains the source code, examples, and project implementations for the book "Hands-on AI Agents" — a practical, code-first guide to building AI agents and, more importantly, putting them into production.

Get the whole plugin

Other skills on atlas-agents.