Tools for creating, auditing, and maintaining Claude Code skills. Includes /create-skill for scaffolding, /review-skill for quality checks, and /audit commands for bulk verification. Use when: building new skills, maintaining skill quality, or forking claude-skills repo.
Installation
Details
Usage
After installing, this skill will be available to your AI coding assistant.
Verify installation:
npx agent-skills-cli listSkill Instructions
name: skill-development description: | Tools for creating, auditing, and maintaining Claude Code skills. Includes /create-skill for scaffolding, /review-skill for quality checks, and /audit commands for bulk verification.
Use when: building new skills, maintaining skill quality, or forking claude-skills repo. user-invocable: true
Skill Development
Tools for creating, auditing, and maintaining Claude Code skills. Essential for anyone building skills or forking this repository.
Slash Commands
/create-skill <name>
Scaffold a new skill from templates.
What it does:
- Validates skill name (lowercase-hyphen-case, max 40 chars)
- Asks about skill type (Cloudflare/AI/Frontend/Auth/Database/Tooling/Generic)
- Copies
templates/skill-skeleton/toskills/<name>/ - Auto-populates name and dates in SKILL.md
- Creates README.md with auto-trigger keywords
- Runs metadata check
- Installs skill for testing
Usage: /create-skill my-new-skill
/review-skill <name>
Quality review and audit of an existing skill.
What it does:
- Checks SKILL.md structure and metadata
- Validates package versions against latest
- Reviews error documentation
- Checks template completeness
- Suggests improvements
Usage: /review-skill cloudflare-worker-base
/audit [name]
Multi-agent audit swarm for parallel skill verification.
What it does:
- Launches parallel agents to audit multiple skills
- Checks versions, metadata, content quality
- Generates consolidated report
Usage:
/audit- Audit all skills/audit cloudflare-*- Audit matching skills
/audit-skill-deep <name>
Deep content validation against official documentation.
What it does:
- Fetches official documentation for the skill's technology
- Compares patterns and versions
- Identifies knowledge gaps or outdated content
- Suggests corrections and updates
Usage: /audit-skill-deep tailwind-v4-shadcn
/deep-audit <name>
Extended deep audit with comprehensive doc scraping.
What it does:
- Scrapes multiple documentation sources
- Caches scraped content in
archive/audit-cache/ - Cross-references with skill content
- Produces detailed accuracy report
Usage: /deep-audit openai-api
/scrape-api <docs-url> [skill-name]
Generate a skill from API documentation URL.
What it does:
- Validates the docs URL is accessible and contains API documentation
- Attempts WebFetch first for static docs
- Falls back to Playwright for JS-heavy sites (readme.io, GitBook)
- Extracts endpoints, authentication, request/response schemas
- Generates complete skill structure with SKILL.md, references/, templates/
- Runs plugin manifest generation
Usage:
/scrape-api https://api.example.com/docs my-api/scrape-api https://rocketdotnet.readme.io/(auto-detects name: rocket-net-api)
Supported Platforms:
- readme.io documentation sites
- Swagger/OpenAPI specs
- GitBook documentation
- Standard HTML API docs
Templates
This skill includes templates for creating new skills:
| Template | Purpose |
|---|---|
templates/skill-skeleton/ | Complete skill directory to copy |
templates/SKILL-TEMPLATE.md | SKILL.md with TODOs to fill |
templates/README-TEMPLATE.md | README.md with keywords section |
templates/skill-metadata-v2.yaml | YAML frontmatter reference |
templates/RESEARCH_FINDINGS_TEMPLATE.md | Research output format |
templates/CONTENT_AUDIT_TEMPLATE.md | Audit checklist |
Using Templates
# Option 1: Use /create-skill (recommended)
/create-skill my-new-skill
# Option 2: Manual copy
cp -r skills/skill-development/templates/skill-skeleton/ skills/my-new-skill/
# Then fill in the TODOs
Quality Standards
Skills should meet these criteria before publishing:
Required
- YAML frontmatter with
nameanddescription - Description includes "Use when" scenarios
- Third-person description style
- Package versions are current
- Templates tested and working
Recommended
- README.md with auto-trigger keywords
- Token efficiency measured (target: 50%+ savings)
- Known issues documented with sources
- Examples include error prevention
Full checklist: See ONE_PAGE_CHECKLIST.md in repo root.
Agents
This skill provides agents for automated skill maintenance:
| Agent | Purpose |
|---|---|
skill-creator | Scaffold new skills with proper structure |
api-doc-scraper | Generate skills from API documentation URLs |
version-checker | Verify package versions are current |
content-accuracy-auditor | Compare skill content vs official docs |
code-example-validator | Validate code examples are syntactically correct |
api-method-checker | Verify documented API methods exist |
doc-validator | Check documentation quality |
bulk-updater | Apply changes across multiple skills |
These agents are in .claude/agents/ (repo-level) and agents/ (skill-bundled) and used by the commands.
Workflow
Creating a New Skill
1. /create-skill my-skill
2. Fill in TODOs in SKILL.md
3. Add templates, references, scripts as needed
4. /review-skill my-skill
5. Fix any issues
6. Test: /plugin install ./skills/my-skill
7. Commit and push
Auditing Skills
1. /audit # Quick check all skills
2. /audit-skill-deep my-skill # Deep check specific skill
3. Review findings
4. Fix issues or document as known limitations
5. Update metadata.last_verified date
Quarterly Maintenance
1. Run: ./scripts/check-all-versions.sh
2. Review VERSIONS_REPORT.md
3. /audit for all skills needing updates
4. Update package versions
5. Test affected skills
6. Commit: "chore: quarterly version updates"
For Forkers
If you've forked this repo to maintain your own skills:
- Install this skill:
/plugin install ./skills/skill-development - Use /create-skill: Creates skills with proper structure
- Use /review-skill: Validates before publishing
- Customize agents: Modify
.claude/agents/for your needs - Pull upstream updates:
git fetch upstream && git merge upstream/main
The tooling is designed to work standalone - you don't need the full repo to use these commands.
More by jezweb
View allInstall and configure shadcn/ui components for React projects. Guides component selection, installation order, dependency management, customisation with semantic tokens, and common UI recipes (forms, data tables, navigation, modals). Use after tailwind-theme-builder has set up the theme infrastructure, when adding components, building forms, creating data tables, or setting up navigation.
Build secure desktop applications with Electron 33, Vite, React, and TypeScript. Covers type-safe IPC via contextBridge, OAuth with custom protocol handlers, native module compatibility (better-sqlite3, electron-store), and electron-builder packaging. Use when building cross-platform desktop apps, implementing OAuth flows in Electron, handling main/renderer process communication, or packaging with code signing. Prevents: NODE_MODULE_VERSION mismatch, hardcoded encryption keys, context isolation bypasses, sandbox conflicts with native modules.
Comprehensive dependency health auditing for JavaScript/TypeScript projects. Run npm audit, detect outdated packages, check for security advisories, and verify license compliance. Prioritises vulnerabilities by severity and provides actionable fix recommendations. Use when: auditing project dependencies, checking for vulnerabilities, updating packages, preparing for release, or investigating "npm audit" warnings. Keywords: audit, vulnerabilities, outdated, security, npm audit, pnpm audit, CVE, GHSA, license.
Generate Drizzle ORM schemas for Cloudflare D1 databases with correct D1-specific patterns. Produces schema files, migration commands, type exports, and DATABASE_SCHEMA.md documentation. Handles D1 quirks: foreign keys always enforced, no native BOOLEAN/DATETIME types, 100 bound parameter limit, JSON stored as TEXT. Use when creating a new database, adding tables, or scaffolding a D1 data layer.
