agentleFS
ZhouYinLong-labZhouYinLong-lab/Dog-Skills/CLAUDE.md

Dog-Skills

CLAUDE.md7 starsChanged 5 months agoMITView on GitHub

A collection of Claude Code skills for AI-assisted development workflows. Each skill lives in its own top-level directory with a SKILL.md file at its root. Pre-built .skill files live in dist/. After creating or copying a skill directory, run: This workflow updates the root README.md in 4 sections: 1. Installation (Method 2) — adds cp -r command 2. Verification — adds entry to ls output 3. Trigger examples table — adds row with example prompts 4. Skills section — adds…

The file

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

7 rules. About 0 are common, 4 are distinctive, and 3 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.

  • 3

    Very specific

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

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

    Create your-skill-name/SKILL.md with YAML frontmatter:

    line 24
  2. Distinctive:

    Run the sync workflow to update the root README (see below).

    line 35
  3. Distinctive:

    After creating or copying a skill directory, run:

    line 39
  4. Very specific:

    Every skill must belong to exactly one category.

    line 77
  5. Distinctive:

    Add the skill entry to the correct category section in alphabetical order

    line 94
  6. Distinctive:

    Update all 4 sub-sections within that category: Installation, Verification, Trigger Table, Skills Detail

    line 95
  7. Very specific:

    Frontmatter: Every SKILL.md starts with ----delimited YAML containing name and description

    line 131

≈ “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

Dog-Skills

A collection of Claude Code skills for AI-assisted development workflows. Each skill lives in its own top-level directory with a SKILL.md file at its root.

Repository structure

skill-directory/
├── SKILL.md          # Required: YAML frontmatter (name, description) + markdown body
├── README.md         # Optional: skill-level documentation
├── LICENSE           # Optional
├── ATTRIBUTIONS.md   # Required for derived works
├── references/       # Optional: supporting materials
├── scripts/          # Optional: automation scripts
└── ...

Pre-built .skill files live in dist/.

Adding a skill

Quick: just the directory

  1. Create your-skill-name/SKILL.md with YAML frontmatter:
    ---
    name: your-skill-name
    description: >
      What this skill does. Trigger keywords: keyword1, keyword2, 中文关键词.
    metadata:
      category: development  # ← REQUIRED: see categories below
    ---
    # Your Skill Title
  2. Run the sync workflow to update the root README (see below).

Workflow: sync skill to README

After creating or copying a skill directory, run:

/workflow sync-skill-readme { skillDir: "your-skill-name/" }

This workflow updates the root README.md in 4 sections:

  1. Installation (Method 2) — adds cp -r command
  2. Verification — adds entry to ls output
  3. Trigger examples table — adds row with example prompts
  4. Skills section — adds full ### \name` — Title` entry

Optional args

ArgDefaultDescription
skillDir(required)Path to skill directory, e.g. "my-skill/"
namefrom SKILL.mdOverride skill name
titlefrom SKILL.mdOverride human-readable title
triggerExamplesderived from descriptionExamples for trigger table: "试试这个" / "try this"
installNamesame as nameDirectory name in ~/.claude/skills/
installCmdauto-generatedInstall command for the Skills section
shortDescriptionfrom SKILL.mdOne-line purpose description
categoryfrom SKILL.md metadataSkill category key (see categories below)

Example: full invocation

/workflow sync-skill-readme {
  skillDir: "tutor/",
  installName: "exam-tutor",
  triggerExamples: "帮我生成复习资料" / "analyze past exams",
  installCmd: "Download `dist/exam-tutor.skill` and drag it into Claude Code."
}

Skill Categories

Every skill must belong to exactly one category. The category is stored in SKILL.md frontmatter under metadata.category.

Category KeyIconNameDescription
thinking🧠Thinking & ResearchDeep research, first-principles thinking, persona distillation, learning methods
development💻DevelopmentCode review, collaboration protocols, site cloning, IDE bridges
design🎨Design & FrontendUI/UX design systems, anti-AI-taste design, animations, HTML→video
content📝Content & WritingAI text humanization, slide decks, diagrams, translation, social content
learning📚Learning & TeachingExam review generators, tutorial generators, course creation
business💼Business & StrategyBusiness diagnostics, decision frameworks, benchmark analysis
tools🔍Tools & DiscoverySkill discovery, reading assistants, IM bridges, utilities

README conventions

The root README.md is organized into category sections (see above). Within each category, skills are listed in alphabetical order. When adding a new skill:

  1. Determine the correct category from the SKILL.md metadata
  2. Add the skill entry to the correct category section in alphabetical order
  3. Update all 4 sub-sections within that category: Installation, Verification, Trigger Table, Skills Detail

Some skills use different source directory names vs install names:

  • tutor/ → installs as exam-tutor
  • nuwa-skill/ → installs as nuwa
  • html-video/ → installs as html-video
  • claude-to-im-skill/ → installs as claude-to-im
  • vercel-labs-skills/ → installs as find-skills
  • last30days-skill/ → installs as last30days
  • ui-ux-pro-max-skill/ → installs as ui-ux-pro-max

Plugin-based skills

Some skills in this repo are wrappers/guides for Claude Code plugins. These use /plugin install instead of cp -r:

  • superpowers/ → /plugin install superpowers@claude-plugins-official
  • planning-with-files/ → /plugin install planning-with-files@claude-plugin-directory
  • code-review/ → /plugin install code-review@claude-plugin-directory
  • code-simplifier/ → /plugin install code-simplifier@claude-plugin-directory
  • webapp-testing/ → /plugin install webapp-testing@claude-plugin-directory
  • mcp-builder/ → /plugin install mcp-builder@claude-plugin-directory
  • academic-research-skills/ → /plugin install academic-research-skills@Imbad0202

These skill directories serve as documentation and trigger-awareness wrappers — when a user's request matches the trigger keywords, Claude Code will suggest the appropriate plugin.

README special sections

The root README.md has two special sections beyond the skill catalog:

  • Usage Workflow: New project setup guide, daily development pipeline, and maintenance cycle
  • Skill Combination Recommendations: Tiered skill bundles for different project types (starter, full pipeline, frontend, large project)

Skill conventions

  • Bilingual: Most skills have Chinese + English names, descriptions, and trigger keywords
  • Attribution: Derived works require ATTRIBUTIONS.md crediting original authors
  • Frontmatter: Every SKILL.md starts with ----delimited YAML containing name and description
  • Licenses: Vary per skill — MIT, MIT-0, Apache-2.0, CC BY-NC 4.0

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.