agentleFS
Sign inStart free
hongphuc5497hongphuc5497/skills/CLAUDE.md

skills

CLAUDE.md2 starsChanged 5 months agoMITView on GitHub

Project context for agents working on this repo. Skills live here; the runtime that loads them lives in ~/.hermes/skills/, ~/.claude/skills/, etc. (out of scope for edits). This is a markdown-only repo. No npm, pnpm, make, or test commands.

  • Reads credentials

The file

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

10 rules. About 0 are common, 5 are distinctive, and 5 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈5

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 5

    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 each SKILL.md under 500 lines.

    line 23
  2. Very specific:

    Never edit dist/ artifacts.

    line 24
  3. Very specific:

    Never commit *-workspace/ or local scratch files.

    line 25
  4. Very specific:

    Never commit secrets or .env.

    line 26
  5. Distinctive:

    Do not commit or push without explicit user request.

    line 28
  6. Very specific:

    Use Conventional Commits (feat:, fix:, chore:, docs:).

    line 29
  7. Distinctive:

    Run validation against any skill you touched before declaring done.

    line 31
  8. Distinctive:

    Never re-read files you just wrote.

    line 35
  9. Distinctive:

    Never re-run commands to "verify" unless outcome was uncertain.

    line 36
  10. Distinctive:

    Don't echo back large blocks of code or file contents unless asked.

    line 37

≈ “Common” is an estimate: at least 1 in 5 of the 26,980 CLAUDE.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
CLAUDE.md

CLAUDE.md — Agent Skills Catalog

Project context for agents working on this repo. Skills live here; the runtime that loads them lives in ~/.hermes/skills/, ~/.claude/skills/, etc. (out of scope for edits).

Commands

  • Validate a skill: check frontmatter, naming, version, structure
  • Scaffold a new skill: mkdir -p skills/<name> && touch skills/<name>/SKILL.md

This is a markdown-only repo. No npm, pnpm, make, or test commands.

Architecture

  • skills/<name>/SKILL.md — required skill definition with YAML frontmatter
  • skills/<name>/references/ — on-demand docs the agent loads when the skill triggers
  • skills/<name>/scripts/ — executable helpers
  • install.sh / remote-install.sh — end-user installers (bash, no deps)

Hard Rules

  1. Bump metadata.version (semver) on every SKILL.md edit.
  2. Quote any frontmatter string containing : # - < > | , & ? !.
  3. Keep each SKILL.md under 500 lines. Split overflow into references/.
  4. Never edit dist/ artifacts.
  5. Never commit *-workspace/ or local scratch files.
  6. Never commit secrets or .env.
  7. Never add Co-Authored-By: Claude to any commit message.
  8. Do not commit or push without explicit user request.
  9. Use Conventional Commits (feat:, fix:, chore:, docs:).
  10. Skill name must exactly match parent directory name (lowercase, hyphens).
  11. Run validation against any skill you touched before declaring done.

Token Efficiency

  • Never re-read files you just wrote. You know the contents.
  • Never re-run commands to "verify" unless outcome was uncertain.
  • Don't echo back large blocks of code or file contents unless asked.
  • Batch related edits into single operations.
  • Skip confirmations like "I'll continue..." Just do it.

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