agentleFS
Sign inStart free
xmtpxmtp/libxmtp/.agents/skills/writing-agent-docs/SKILL.md

writing-agent-docs

Skill127 starsChanged 11 days agoMITView on GitHub

Use when creating, editing, or reviewing AGENTS.md, CLAUDE.md, skills, or linked agent guidance.

The file

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:

    Every token should change an agent's decisions.

    line 8
  2. Distinctive:

    Keep AGENTS.md to directory purpose, essential commands, and consequential constraints.

    line 10
  3. Distinctive:

    Put task-specific procedures, troubleshooting, and examples in focused docs or skills.

    line 13
  4. Distinctive:

    Document the general constraint, not the story of one failure.

    line 19
  5. Distinctive:

    Verify commands and claims against current files.

    line 24
  6. Distinctive:

    Avoid history searches unless current sources leave a material ambiguity.

    line 26

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 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

Writing agent docs

Every token should change an agent's decisions.

  • Keep AGENTS.md to directory purpose, essential commands, and consequential constraints. Shared rules belong in the nearest common parent; do not repeat them in children. CLAUDE.md stays a pointer: @AGENTS.md.
  • Put task-specific procedures, troubleshooting, and examples in focused docs or skills. Link them with a clear trigger for when to read more. Prune before moving; relocating bloat does not fix it.
  • Reuse owning specs, Just recipes, and maintained guides. Avoid copied requirements, command catalogs, source inventories, line numbers, generic advice, history, and temporary status.
  • Document the general constraint, not the story of one failure. Prefer enforcing a rule in tooling when practical; avoid permanent instructions for one-off workarounds.
  • Preserve critical constraints, prerequisites, and platform exceptions. Shortening or moving text must not weaken an obligation.
  • Verify commands and claims against current files. Follow each new link: its target must contain accurate, useful guidance, not merely exist. Avoid history searches unless current sources leave a material ambiguity.
  • Review the diff for lost meaning. Each removed rule should be redundant, obsolete, inherited, or discoverable through a relevant link. Run the applicable documentation checks and report any that could not run.

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.

Check your setup, free