agentleFS
Sign inStart free
ArvindChanderArvindChander/skill-lab/AGENTS.md

skill-lab

AGENTS.md1 starsChanged 4 months agoMITView on GitHub

This repository publishes portable Agent Skills. Apply these instructions to every file in the repository. Run these from the repository root: Report the checks actually run and any unverified behavior.

  • Where files are headingNew AGENTS.md files have more rules about keeping secrets out of the code, and fewer about code style.

    Of the AGENTS.md files we read, the 31,774 first seen in the last six months, against the older ones: the kind of rule that gained most share, and the kind that lost most. About files like this in general, not this one, and an estimate, since a kind is a pattern in the words.

  • In this repository5 other files it gives its agents.

The file

We split this AGENTS.md into the rules it gives an agent

10 rules. About 0 are common, 3 are distinctive, and 7 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈3

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 7

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  • could be automatic: a setting could enforce it
  1. Very specific:

    Keep SKILL.md frontmatter limited to name and description for broad compatibility.

    line 8
  2. Distinctive:

    Keep the instructions agent-neutral.

    line 9
  3. Very specific:

    Treat agents/openai.yaml as optional interface metadata, not as a dependency of the skill.

    line 10
  4. Very specific:

    Put detailed, conditional guidance in references/ and link to it directly from SKILL.md.

    line 11
  5. Very specific:

    Never edit dist/*.skill by hand.

    line 12
  6. Distinctive:

    Do not add silent package installation, persistent initialization, destructive actions, or remote transmission.

    line 17
  7. Very specific:

    When adding or changing an upstream integration, review its current behavior and license, then update UPSTREAMS.md.

    line 18
  8. Very specific:

    Update evals/cases.json when routing behavior changes and add a dated forward-test result for material workflow changes.

    line 19
  9. Very specific:

    Update README.md and MIGRATION.md when a public name or replacement path changes.

    line 20
  10. Distinctive:

    Run these from the repository root:

    line 24

≈ “Common” is an estimate: at least 1 in 5 of the 40,246 AGENTS.md 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
AGENTS.md

Repository guidance

This repository publishes portable Agent Skills. Apply these instructions to every file in the repository.

Canonical content

  • First-level directories containing SKILL.md are the canonical skill sources.
  • Keep SKILL.md frontmatter limited to name and description for broad compatibility.
  • Keep the instructions agent-neutral. A skill may name optional integrations, but its core workflow must remain useful without them.
  • Treat agents/openai.yaml as optional interface metadata, not as a dependency of the skill.
  • Put detailed, conditional guidance in references/ and link to it directly from SKILL.md.
  • Never edit dist/*.skill by hand. Regenerate archives after source changes.

Change requirements

  • Make trigger descriptions state both positive use cases and important exclusions.
  • Do not add silent package installation, persistent initialization, destructive actions, or remote transmission.
  • When adding or changing an upstream integration, review its current behavior and license, then update UPSTREAMS.md.
  • Update evals/cases.json when routing behavior changes and add a dated forward-test result for material workflow changes.
  • Update README.md and MIGRATION.md when a public name or replacement path changes.

Required checks

Run these from the repository root:

python3 scripts/validate_skills.py
python3 scripts/package_skills.py
python3 scripts/validate_packages.py
git diff --exit-code -- dist

Report the checks actually run and any unverified behavior.

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