Agent Skills
Agent Skills Directory
Discover and install AI agent skills from the skills.sh ecosystem.
86,653
Total Skills
2.8M
Total Installs
#1 Most Installed
Install
npx skills add vercel-labs/skills --skill find-skills 458.2K
Installs
Top Skills
Ranked by total installs across the ecosystem.
npx skills add vercel-labs/agent-skills --skill vercel-react-best-practices Safe npx skills add vercel-labs/agent-skills --skill web-design-guidelines Medium npx skills add anthropics/skills --skill frontend-design Safe npx skills add remotion-dev/skills --skill remotion-best-practices Medium npx skills add sleekdotdesign/agent-skills --skill sleek-design-mobile-apps Safe npx skills add vercel-labs/agent-skills --skill vercel-composition-patterns Safe npx skills add anthropics/skills --skill skill-creator Safe npx skills add vercel-labs/agent-skills --skill vercel-react-native-skills Safe npx skills add coreyhaines31/marketingskills --skill seo-audit Medium npx skills add squirrelscan/skills --skill audit-website Medium npx skills add anthropics/skills --skill pdf Critical npx skills add supabase/agent-skills --skill supabase-postgres-best-practices Safe npx skills add coreyhaines31/marketingskills --skill copywriting Safe npx skills add vercel-labs/next-skills --skill next-best-practices Safe npx skills add anthropics/skills --skill pptx Safe npx skills add anthropics/skills --skill docx Safe npx skills add toolshell/skills --skill agent-browser Critical npx skills add toolshell/skills --skill qwen-image-2-pro npx skills add toolshell/skills --skill qwen-image-2 npx skills add anthropics/skills --skill xlsx Safe npx skills add coreyhaines31/marketingskills --skill marketing-psychology Safe npx skills add anthropics/skills --skill webapp-testing Safe npx skills add better-auth/skills --skill better-auth-best-practices Safe npx skills add coreyhaines31/marketingskills --skill programmatic-seo Safe npx skills add coreyhaines31/marketingskills --skill content-strategy Medium npx skills add anthropics/skills --skill mcp-builder Medium npx skills add coreyhaines31/marketingskills --skill product-marketing-context Safe npx skills add coreyhaines31/marketingskills --skill marketing-ideas Safe npx skills add coreyhaines31/marketingskills --skill social-content Medium npx skills add coreyhaines31/marketingskills --skill copy-editing Safe npx skills add coreyhaines31/marketingskills --skill pricing-strategy Safe npx skills add anthropics/skills --skill canvas-design Safe npx skills add expo/skills --skill building-native-ui Safe npx skills add coreyhaines31/marketingskills --skill page-cro Safe npx skills add coreyhaines31/marketingskills --skill analytics-tracking Safe npx skills add coreyhaines31/marketingskills --skill launch-strategy Safe npx skills add coreyhaines31/marketingskills --skill schema-markup Safe npx skills add roin-orca/skills --skill fun-brainstorming Safe npx skills add coreyhaines31/marketingskills --skill competitor-alternatives Medium npx skills add coreyhaines31/marketingskills --skill email-sequence Safe npx skills add coreyhaines31/marketingskills --skill paid-ads Safe npx skills add coreyhaines31/marketingskills --skill onboarding-cro Safe npx skills add coreyhaines31/marketingskills --skill form-cro Safe npx skills add coreyhaines31/marketingskills --skill referral-program Safe npx skills add coreyhaines31/marketingskills --skill free-tool-strategy Safe npx skills add coreyhaines31/marketingskills --skill signup-flow-cro Safe npx skills add coreyhaines31/marketingskills --skill paywall-upgrade-cro Safe npx skills add coreyhaines31/marketingskills --skill ab-test-setup Safe npx skills add google-labs-code/stitch-skills --skill react:components Top Sources
Organizations and authors with the most installed skills.
What are Agent Skills?
Agent skills are instruction packages that teach AI coding agents how to perform specific tasks. Unlike plugins or extensions that add runtime code, skills provide structured knowledge — best practices, coding patterns, framework conventions, and workflow instructions — that agents like Claude Code, Cursor, and Windsurf use to produce better, more contextual output. Think of them as expert playbooks: a React skill teaches your agent how to structure components, manage state, and follow hooks conventions, while a security skill teaches it to spot and prevent common vulnerabilities.
How do skills work?
Each skill is a text-based instruction file stored in a GitHub repository. When installed, it becomes part of your project's context that the AI agent reads before generating code. Skills are versioned, open-source, and community-maintained. The skills.sh registry indexes skills from across GitHub, tracks install counts, and runs automated security audits through multiple scanners (Socket, Snyk, ATH) to ensure packages are safe to use.
How to install a skill
Installation is a single command. Run the install command shown on any skill card (e.g., npx skills add owner/repo --skill skill-name) in your project directory. The skill file is downloaded and placed in your project. No build step, no configuration — your AI agent picks it up automatically on the next interaction. You can install multiple skills, update them individually, or remove them at any time.
Browse by source
Every skill is linked to a GitHub repository. Organizations like Vercel, Anthropic, and independent developers publish skill collections covering frameworks (React, Next.js, Vue), languages (TypeScript, Python, Rust), tools (Docker, Terraform, Kubernetes), and domains (accessibility, performance, security). Browse by source to find curated skill sets from authors you trust.
Why use skills?
Without skills, AI agents rely on their general training data, which may be outdated or lack project-specific conventions. Skills fill this gap: they encode the latest framework patterns, your organization's coding standards, and domain expertise directly into the agent's context. Teams report fewer code review iterations, more consistent output, and faster onboarding when agents are equipped with the right skills.
Security and trust
Every skill indexed in the registry undergoes automated security analysis through multiple independent scanners. Risk levels (Safe, Medium, Critical) are displayed on each skill card so you can make informed decisions. Since skills are plain text instructions hosted on GitHub, you can always review the exact content before installing. There's no hidden code execution — skills are read-only context for your AI agent.
Data on this page is sourced from the skills.sh agent skills registry. Skills and install counts are updated regularly.
Made in Europe Chat with 100+ AI Models in one App.
Use Claude, ChatGPT, Gemini alongside with EU-Hosted Models like Deepseek, GLM-5, Kimi K2.5 and many more.
