agentleFS
schweppsschwepps/skills/CLAUDE.md

skills

CLAUDE.md16 starsChanged 10 months agoMITView on GitHub

This is a Claude Code skills marketplace repository containing documentation-based skills that extend Claude Code with specialized workflows. Skills are pure markdown - no build system, dependencies, or executable code. Critical: The description field determines when Claude activates the skill: - Use third-person phrasing: "This skill should be used when..." - Include exact phrases users would say to trigger it - Be specific and concrete, not vague After frontmatter, include: - When to Use: Clear activation criteria - Workflow/Methodology: Phased…

The file

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

5 rules. About 0 are common, 2 are distinctive, and 3 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    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. Distinctive:

    Use third-person phrasing: "This skill should be used when..."

    line 52
  2. Very specific:

    Place supporting materials in references/:

    line 67
  3. Very specific:

    Add new skills to .claude-plugin/marketplace.json:

    line 75
  4. Very specific:

    Only name, source, and description are valid plugin fields

    line 100
  5. Distinctive:

    Checklists: Use □ for unchecked items

    line 122

≈ “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.

Project Overview

This is a Claude Code skills marketplace repository containing documentation-based skills that extend Claude Code with specialized workflows. Skills are pure markdown - no build system, dependencies, or executable code.

Project Structure

skills/
├── .claude-plugin/
│   ├── marketplace.json        # Full marketplace (all skills)
│   ├── marketplace-seo.json    # SEO skills only
│   └── marketplace-music.json  # Music skills only
├── scripts/
│   └── build-skills.sh         # Build script for packaging
├── dist/                       # Generated packages (after build)
├── skill-name/
│   ├── SKILL.md                # Required: Main skill definition
│   └── references/             # Optional: Supporting documentation
│       └── *.md files
└── CLAUDE.md

Creating a New Skill

1. Create Skill Directory

mkdir -p skill-name/references
touch skill-name/SKILL.md

2. SKILL.md Frontmatter

---
name: skill-name
description: This skill should be used when the user asks to "specific phrase 1", "specific phrase 2", or "specific phrase 3". Be concrete about trigger conditions.
license: MIT
metadata:
  author: Schwepps
  version: "1.0.0"
  category: marketing|music|development|etc
  tags: tag1, tag2, tag3
---

Critical: The description field determines when Claude activates the skill:

  • Use third-person phrasing: "This skill should be used when..."
  • Include exact phrases users would say to trigger it
  • Be specific and concrete, not vague

3. SKILL.md Content Structure

After frontmatter, include:

  • When to Use: Clear activation criteria
  • Workflow/Methodology: Phased approach with specific steps
  • Output Format: Template showing expected deliverable structure
  • Scoring Guide (for audits): Quantitative evaluation criteria
  • Quick Version: Abbreviated workflow for rapid execution

4. Reference Files

Place supporting materials in references/:

  • Checklists with verification steps
  • Templates and examples
  • Tool commands and configurations
  • Domain-specific guides

Marketplace Registration

Add new skills to .claude-plugin/marketplace.json:

{
  "owner": {
    "name": "schwepps",
    "url": "https://github.com/schwepps"
  },
  "name": "skills",
  "version": "1.0.0",
  "description": "Repository description",
  "license": "MIT",
  "plugins": [
    {
      "name": "skill-name",
      "source": "./skill-name",
      "description": "One-line description of the skill"
    }
  ]
}

Important schema requirements:

  • owner must be an object with name and url fields
  • source paths must start with ./
  • Only name, source, and description are valid plugin fields

Also add to the appropriate category marketplace (marketplace-seo.json, marketplace-music.json, etc.).

Conventions

AspectStandard
Namingkebab-case for directories and skill names
LicenseMIT
AuthorSchwepps
VersionSemver, start at 1.0.0
Categoriesseo, music (extensible)
TagsLowercase, comma-separated

Content Patterns

Established patterns from existing skills:

  • Phased workflows: Phase 1, Phase 2, etc. with clear deliverables
  • Scoring rubrics: 0-100 scale with tiers (90-100 Excellent, 70-89 Good, 50-69 Needs Work, <50 Critical)
  • Status indicators: [Good], [Needs Work], [Critical]
  • Checklists: Use □ for unchecked items
  • Priority actions: Immediate → Short-term → Ongoing
  • Output tables: Markdown tables for structured data

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.