agentleFS
Sign inStart free
ilchemlailchemla/agent-skills/CLAUDE.md

agent-skills

CLAUDE.md4 starsChanged 23 days agoMITView on GitHub

Agent skills repository for Claude Code. Skills are stored in skills/ with each skill in its own directory containing a SKILL.md manifest and optional references/ directory for supporting documentation. Each skill requires YAML frontmatter: Key Requirements: - name: Kebab-case identifier - description: Must include "Use when" clause describing invocation scenarios - Description should clearly specify trigger conditions for Claude Code to auto-invoke the skill qodo-pr-resolver: Wait, Analyze, Confirm, Execute workflow for Qodo PR review findings. Reads them as structured JSON…

The file

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

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

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈0

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 3

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Add SKILL.md with required frontmatter and documentation

    line 47
  2. Very specific:

    Add references/ subdirectory for supporting docs if needed

    line 49
  3. Very specific:

    Commit with conventional commit format: feat: add <skill-name> skill

    line 50

≈ “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

This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.

Overview

Agent skills repository for Claude Code. Skills are stored in skills/ with each skill in its own directory containing a SKILL.md manifest and optional references/ directory for supporting documentation.

Repository Structure

skills/
  <skill-name>/
    SKILL.md          # Skill manifest with YAML frontmatter
    references/       # Optional supporting documentation

Skill Manifest Format (SKILL.md)

Each skill requires YAML frontmatter:

---
name: skill-name
description: >
  Brief description of when to use this skill.
  Use when [specific scenarios].
---

Key Requirements:

  • name: Kebab-case identifier
  • description: Must include "Use when" clause describing invocation scenarios
  • Description should clearly specify trigger conditions for Claude Code to auto-invoke the skill

Current Skills

qodo-pr-resolver: Wait, Analyze, Confirm, Execute workflow for Qodo PR review findings. Reads them as structured JSON through the Qodo CLI 2.0+ (falls back to GitHub comments), works only on a review at the PR head, fixes one commit per finding, and records outcomes in Qodo (dismiss / mark-implemented).

panel: End-of-task review by 3-6 expert subagents picked for the work (code, docs, plans, messaging, slides, diagrams). Renders visual artifacts to images before review. Verifies findings, returns a P0/P1/P2 report with a ship / fix-first / rethink verdict, applies fixes only after approval. Can add a private "voice" of a real person, built from their own messages and meetings, that reacts to the other reviewers. User state and voices live in ~/.claude/panel/, never in the skill folder.

tldr: Bottom line of a long reply, plain-words explanations, and open questions one at a time. Optional always-on mode through a UserPromptSubmit hook (off by default). Learnings live in ~/.claude/tldr/.

Adding New Skills

  1. Create skills/<skill-name>/ directory
  2. Add SKILL.md with required frontmatter and documentation
  3. Include detailed workflow, prerequisites, and usage examples
  4. Add references/ subdirectory for supporting docs if needed
  5. Commit with conventional commit format: feat: add <skill-name> skill

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