Awesome AI Skills: The Curated Agent Skills Directory

Discover, search, and install the best open AI agent skills from GitHub—organized, up-to-date, and ready to use.

Built by RanBOT Labs

Features

• Multi-Source Scraper: Collects agent skills from 6+ GitHub repositories
• Modern Website: Next.js 16 + Tailwind CSS frontend with search, filtering, and detailed skill pages
• 300+ Skills: Curated from top AI agent skill repositories
• Categories: Skills organized by category (Development, AI & Agents, Business, Creative, etc.)
• Search & Filter: Find skills by name, description, tags, or category
• Automated Updates: GitHub Actions workflow for scheduled scraping

Data Sources

Repository	Stars	Description
obra/superpowers	35k+	Agentic skills framework & development methodology
anthropics/skills	52k+	Official Anthropic Claude skills
ComposioHQ/awesome-claude-skills	25k+	Community Claude AI skills
OpenHands/OpenHands	67k+	AI-driven development skills
Prat011/awesome-llm-skills	700+	LLM and AI Agent skills collection
sickn33/antigravity-awesome-skills	3.6k+	253+ agentic skills for Claude Code, Gemini CLI, Cursor & more

Install Skills

Install skills for use with Claude Code, Cursor, Codex, Gemini CLI, and 36+ other AI agents.

Web Interface: Visit the Install page for an interactive installation guide.

Method 1: Using skills.sh CLI (Recommended)

The easiest way to install skills using the skills.sh ecosystem:

# Install all skills to all your agents
npx skills add ranbot-ai/awesome-skills --all

# List available skills first
npx skills add ranbot-ai/awesome-skills --list

# Install specific skills
npx skills add ranbot-ai/awesome-skills --skill code-review --skill docker

# Install to specific agents only
npx skills add ranbot-ai/awesome-skills -a claude-code -a cursor

# Install globally (available across all projects)
npx skills add ranbot-ai/awesome-skills -g

Supported Agents: Claude Code, Cursor, Codex, Gemini CLI, GitHub Copilot, Cline, OpenCode, Windsurf, Roo Code, and 36+ more.

Method 2: One-liner Script

Install skills directly to ~/.claude/skills:

# Install all skills
curl -fsSL https://raw.githubusercontent.com/ranbot-ai/awesome-skills/main/scripts/install-skills.sh | bash

# Install specific skills
curl -fsSL https://raw.githubusercontent.com/ranbot-ai/awesome-skills/main/scripts/install-skills.sh | bash -s -- code-review docker kubernetes

# Install all skills from a source
curl -fsSL https://raw.githubusercontent.com/ranbot-ai/awesome-skills/main/scripts/install-skills.sh | bash -s -- --source anthropic

Method 3: From Local Clone

See Using the Install Script below for advanced options.

Installed Directory Structure

Skills are installed following the official Agent Skills specification:

~/.claude/skills/
├── code-review/
│   └── SKILL.md          # Skill instructions with frontmatter
├── docker/
│   └── SKILL.md
├── kubernetes/
│   └── SKILL.md
└── ...

Each skill folder can also contain optional subdirectories:

• scripts/ - Executable scripts for the skill
• references/ - Reference documentation
• assets/ - Images or other assets

Using the Install Script

# Clone the repo first
git clone https://github.com/ranbot-ai/awesome-skills.git
cd awesome-skills

# List available skills
./scripts/install-skills.sh --list

# Install all skills
./scripts/install-skills.sh

# Install specific skills
./scripts/install-skills.sh code-review docker kubernetes

# Install from local data (after scraping)
./scripts/install-skills.sh --local ./data

# Search for skills
./scripts/install-skills.sh --search docker

# Install all skills from a source (anthropic, composio, superpowers, openhands, awesome-llm, antigravity)
./scripts/install-skills.sh --source anthropic

# Install skills by category
./scripts/install-skills.sh --category "Development"

# Show skill info
./scripts/install-skills.sh --info code-review

# Update existing skills
./scripts/install-skills.sh --update

# Uninstall a skill
./scripts/install-skills.sh --uninstall code-review

# Remove all installed skills
./scripts/install-skills.sh --clean

Install Script Options

Option	Description
--list, -l	List all available skills
--all, -a	Install all skills (default)
--local, -L PATH	Use local data directory
--update, -u	Update/overwrite existing skills
--source, -s NAME	Install skills from a specific source
--category, -c NAME	Install skills matching a category
--search, -S TERM	Search skills by name/description
--info, -i SKILL	Show detailed info about a skill
--uninstall SKILL	Remove a specific skill
--clean	Remove all installed skills

skills.sh CLI Options

When using npx skills add, these options are available:

Option	Description
--list, -l	List available skills without installing
--skill, -s NAME	Install specific skills by name
--agent, -a NAME	Target specific agents (claude-code, cursor, codex, etc.)
--global, -g	Install to user directory instead of project
--yes, -y	Skip confirmation prompts (CI/CD friendly)
--all	Install all skills to all agents

Environment Variables

Variable	Description
CLAUDE_SKILLS_DIR	Override skills directory (default: ~/.claude/skills)

---

Quick Start (Development)

1. Install Dependencies

npm run setup

2. Run the Scraper

# Just scrape
npm run scrape

# Build scraper first, then scrape
npm run scrape:build

This will fetch the latest skills from GitHub and save them to the data/ directory.

Note: For higher rate limits (5000 req/hour vs 60), set a GitHub token:

export GITHUB_TOKEN=your_github_token
npm run scrape

3. Start the Development Server

npm run dev

Visit http://localhost:3000 to see the website.

4. Build for Production

npm run build
npm run start

Scripts

Command	Description
npm run setup	Install all dependencies
npm run scrape	Run the scraper to fetch skills
npm run scrape:build	Build and run the scraper
npm run update-skills	Scrape and copy data to web
npm run dev	Start development server
npm run build	Build for production
npm run start	Start production server
npm run deploy	Scrape, copy data, commit & push
npm run deploy:build	Full deploy with web build

Automated Scraping

Using the Shell Script

# Just scrape and update data
./scripts/scrape-and-deploy.sh

# Scrape and push to remote
./scripts/scrape-and-deploy.sh --push

# Scrape, push, and build web app
./scripts/scrape-and-deploy.sh --push --build

Using GitHub Actions

The project includes a GitHub Actions workflow that runs daily at 6 AM UTC:

• Automatic: Scheduled daily scraping
• Manual: Trigger from GitHub Actions tab
• Commits: Automatically commits and pushes updates

See .github/workflows/scrape-skills.yml for configuration.

Project Structure

awesome-skills/
├── scraper/                # Node.js skill scraper
│   ├── src/
│   │   ├── index.ts        # Main entry point
│   │   ├── types.ts        # TypeScript types
│   │   ├── github.ts       # GitHub API utilities
│   │   ├── processor.ts    # Data processing
│   │   └── scrapers/       # Source-specific scrapers
│   │       ├── anthropic.ts
│   │       ├── antigravity.ts
│   │       ├── awesome-llm.ts
│   │       ├── composio.ts
│   │       ├── openhands.ts
│   │       └── superpowers.ts
│   └── package.json
├── web/                    # Next.js frontend
│   ├── src/
│   │   ├── app/            # Next.js app router
│   │   │   ├── page.tsx    # Homepage
│   │   │   ├── about/      # About page
│   │   │   ├── install/    # Install guide page
│   │   │   └── skill/      # Skill detail pages
│   │   ├── components/     # React components
│   │   ├── lib/            # Utility functions
│   │   └── types/          # TypeScript types
│   └── package.json
├── scripts/                # Automation scripts
│   ├── scrape-and-deploy.sh  # Scrape and deploy to repo
│   ├── build-static.sh       # Build static site
│   └── install-skills.sh     # Install skills locally
├── .github/
│   └── workflows/
│       └── scrape-skills.yml
├── data/                   # Scraped skill data
│   ├── skills.json         # All skills data
│   ├── search-index.json   # Search index
│   └── skills/             # Individual skill files
└── package.json            # Root package.json

Website Features

Homepage

• Hero section with animated gradient
• Source cards with GitHub stars
• Search bar with instant filtering
• Category cards for browsing by category
• Skills grid with hover effects
• Filter by category and source

Install Page

• Interactive installation guide
• One-click copy commands
• Multiple installation methods (one-liner, selective, local clone)
• Source-specific installation
• Full command reference
• Requirements and configuration

About Page

• Agent Skills specification overview
• Compatible tools list
• Links to official documentation

Skill Detail Page

• Full documentation display
• Tags and metadata
• Use cases (when available)
• Related skills
• Quick links to GitHub
• Scraped date tracking

Technologies

Scraper

• TypeScript - Type-safe JavaScript
• gray-matter - Parse YAML frontmatter
• GitHub API - Fetch repository contents

Website

• Next.js 16 - React framework with App Router
• Tailwind CSS - Utility-first CSS
• Lucide React - Beautiful icons
• TypeScript - Type safety

Adding New Sources

To add a new skill source:

1. Create a scraper in scraper/src/scrapers/:

import type { Skill, GitHubRepo } from '../types.js';

export const NEW_REPO: GitHubRepo = {
  owner: 'owner',
  repo: 'repo',
  branch: 'main',
  skillsPath: 'skills',
};

export async function scrapeNewSource(): Promise<Skill[]> {
  // Implementation
}

2. Add the source type to scraper/src/types.ts

3. Update scraper/src/processor.ts to include the new source

4. Import and call the scraper in scraper/src/index.ts

5. Update web/src/lib/skills.ts with source label and color

6. Update web/src/components/Footer.tsx with source link

Environment Variables

Variable	Description
GITHUB_TOKEN	GitHub token for higher API rate limits
GH_TOKEN	Alternative GitHub token variable

License

MIT

---

RanBOT Labs - Building intelligent tools for the AI era.