agentleFS
ming-kangming-kang/Skills/AGENTS.md

Skills

AGENTS.md3 starsChanged 3 days agoMITView on GitHub

This repository is a personal repository for agent Skills. When changing skill behavior: Run the frontmatter validator after any SKILL.md frontmatter change; it catches the YAML and spec errors that make installers reject a skill. npm run validate:skills is the same command. For Visualize asset or helper changes, also run the relevant build, browser review, and regression checks documented in docs/README.visualize.md. Before proposing a commit, review every changed file. Do not commit or push unless the user asks.

  • 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 32,671 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 repository3 other files it gives its agents.

The file

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

12 rules. About 0 are common, 4 are distinctive, and 8 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈4

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 8

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Keep each skill's supporting files inside that skill directory.

    line 8
  2. Very specific:

    Keep development dependencies at the repository root and generated review artifacts under .artifacts/.

    line 11
  3. Very specific:

    Put templates and long reference material under <skill>/references/.

    line 12
  4. Very specific:

    Put runtime helper scripts under <skill>/scripts/.

    line 13
  5. Very specific:

    Put reusable assets under <skill>/assets/.

    line 14
  6. Very specific:

    Put user-facing documentation under docs/ as README.<name>.md.

    line 15
  7. Distinctive:

    When changing skill behavior:

    line 19
  8. Very specific:

    Update the relevant SKILL.md.

    line 21
  9. Very specific:

    Update that skill's references/*.md template if generated artifacts change.

    line 22
  10. Distinctive:

    Check whether README.md needs user-facing documentation updates.

    line 23
  11. Very specific:

    Run the frontmatter validator after any SKILL.md frontmatter change.

    line 34
  12. Distinctive:

    Before proposing a commit, review every changed file.

    line 38

≈ “Common” is an estimate: at least 1 in 5 of the 41,581 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

Repo Contracts

This repository is a personal repository for agent Skills.

Structure Rules

  • Each usable skill is a top-level directory containing SKILL.md.
  • Keep each skill's supporting files inside that skill directory.
  • The entire skill directory is distributed as one unit. Keep only runtime instructions, runtime helpers, and reusable assets inside it; repository-only build sources, tests, and browser tooling belong under tools/<skill>/.
  • Repository-wide tooling that applies to every skill (for example tools/quick_validate.py, the frontmatter validator) lives directly under tools/.
  • Keep development dependencies at the repository root and generated review artifacts under .artifacts/; neither belongs in a distributable skill folder.
  • Put templates and long reference material under <skill>/references/.
  • Put runtime helper scripts under <skill>/scripts/.
  • Put reusable assets under <skill>/assets/.
  • Put user-facing documentation under docs/ as README.<name>.md.

Change Checklist

When changing skill behavior:

  • Update the relevant SKILL.md.
  • Update that skill's references/*.md template if generated artifacts change.
  • Check whether README.md needs user-facing documentation updates.
  • Confirm skill instructions and templates agree on filenames, statuses, frontmatter fields, and archive layout.

Validation

git status --short
git diff --check
python tools/quick_validate.py   # SKILL.md frontmatter of every skill (needs PyYAML)

Run the frontmatter validator after any SKILL.md frontmatter change; it catches the YAML and spec errors that make installers reject a skill. npm run validate:skills is the same command.

For Visualize asset or helper changes, also run the relevant build, browser review, and regression checks documented in docs/README.visualize.md.

Before proposing a commit, review every changed file. Do not commit or push unless the user asks.

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.