New here? This guide will help you supercharge your AI Agent in 5 minutes.
AI Agents (like Claude Code, Gemini, Cursor) are smart, but they lack specific knowledge about your tools. Skills are specialized instruction manuals (markdown files) that teach your AI how to perform specific tasks perfectly, every time.
Analogy: Your AI is a brilliant intern. Skills are the SOPs (Standard Operating Procedures) that make them a Senior Engineer.
Don't panic about the 626+ skills. You don't need them all at once. We have curated Starter Packs to get you running immediately.
You install the full repo once (npx or clone); Starter Packs are curated lists to help you pick which skills to use by role (e.g. Web Wizard, Hacker Pack)—they are not a different way to install.
Option A — npx (easiest):
npx antigravity-awesome-skillsThis clones to ~/.agent/skills by default. Use --cursor, --claude, --gemini, or --codex to install for a specific tool, or --path <dir> for a custom location. Run npx antigravity-awesome-skills --help for details.
If you see a 404 error, use: npx github:sickn33/antigravity-awesome-skills
Option B — git clone:
# Universal (works for most agents)
git clone https://github.com/sickn33/antigravity-awesome-skills.git .agent/skillsFind the bundle that matches your role (see BUNDLES.md):
| Persona | Bundle Name | What's Inside? |
|---|---|---|
| Web Developer | Web Wizard |
React Patterns, Tailwind mastery, Frontend Design |
| Security Engineer | Hacker Pack |
OWASP, Metasploit, Pentest Methodology |
| Manager / PM | Product Pack |
Brainstorming, Planning, SEO, Strategy |
| Everything | Essentials |
Clean Code, Planning, Validation (The Basics) |
Once installed, just talk to your AI naturally.
"Use @brainstorming to help me design a new login flow."
What happens: The AI loads the brainstorming skill, asks you structured questions, and produces a professional spec.
"Run @lint-and-validate on this file and fix errors."
What happens: The AI follows strict linting rules defined in the skill to clean your code.
"Use @api-security-best-practices to review my API endpoints."
What happens: The AI audits your code against OWASP standards.
| Tool | Status | Path |
|---|---|---|
| Claude Code | ✅ Full Support | .claude/skills/ |
| Gemini CLI | ✅ Full Support | .gemini/skills/ |
| Codex CLI | ✅ Full Support | .codex/skills/ |
| Antigravity | ✅ Native | .agent/skills/ |
| Cursor | ✅ Native | .cursor/skills/ |
| Copilot | Manual copy-paste |
We classify skills so you know what you're running:
- 🟣 Official: Maintained by Anthropic/Google/Vendors (High Trust).
- 🔵 Safe: Community skills that are non-destructive (Read-only/Planning).
- 🔴 Risk: Skills that modify systems or perform security tests (Authorized Use Only).
Check the Skill Catalog for the full list.
Q: Do I need to install all 626 skills? A: You clone the whole repo once; your AI only reads the skills you invoke (or that are relevant), so it stays lightweight. Starter Packs in BUNDLES.md are curated lists to help you discover the right skills for your role—they don't change how you install.
Q: Can I make my own skills? A: Yes! Use the @skill-creator skill to build your own.
Q: Is this free? A: Yes, MIT License. Open Source forever.