agentleFS
zekhoizekhoi/claude/.claude/skills/kit-documentation/SKILL.md

kit-documentation

Skill0 starsChanged 20 days agoMITView on GitHub

Write or update project documentation such as READMEs, setup and usage guides, and API references so it matches what the code actually does, with every command and example verified. Use when the user asks to write, update, or fix documentation, a README, or usage instructions. Do not use for code comments during feature work, changelogs for releases, or explaining code to the user in chat.

We split this skill into the rules it gives an agent

6 rules. About 0 are common, 6 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈6

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Read the code, configuration, and scripts the documentation describes: entry points, CLI arguments, environment variables, public APIs, and …

    line 12
  2. Distinctive:

    Read the existing documentation and note what is missing, outdated, or contradicted by the code.

    line 13
  3. Distinctive:

    Keep it concise. Link to existing documents instead of duplicating them.

    line 21
  4. Distinctive:

    Run every command you document where it is safe to do so, and check examples against the code.

    line 25
  5. Distinctive:

    Check that relative links and referenced files exist.

    line 27
  6. Distinctive:

    Use: "Write a README for this project." "The setup docs are out of date.

    line 35

≈ “Common” is an estimate: at least 1 in 5 of the 308,232 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Documentation

Documentation that disagrees with the code is worse than none. Write only what you have verified.

1. Find the source of truth

  • Read the code, configuration, and scripts the documentation describes: entry points, CLI arguments, environment variables, public APIs, and build and test commands.
  • Read the existing documentation and note what is missing, outdated, or contradicted by the code.
  • Identify the audience (users, contributors, operators) and keep the existing structure and tone unless asked to change them.

2. Write

  • Lead with what the reader needs first: what it is, how to install or set it up, and how to use it.
  • Show real, copyable commands and examples taken from the project. Never invent flags, options, endpoints, or output.
  • Explain prerequisites and supported versions as the project defines them.
  • Keep it concise. Link to existing documents instead of duplicating them.

3. Verify

  • Run every command you document where it is safe to do so, and check examples against the code. Do not run commands that deploy, publish, send messages, or modify shared systems.
  • Mark anything you could not verify, and tell the user, rather than presenting it as fact.
  • Check that relative links and referenced files exist.

4. Report

Summarize what was added or changed, what was verified and how, and anything left unverified or needing the maintainer's input.

Examples

  • Use: "Write a README for this project." "The setup docs are out of date; fix them."
  • Skip: "Write release notes for 2.0." "Explain how this function works." (answer in chat)

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.

Reviews can't be read right now.

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.