One CLI. Every Source. Every Agent. Zero Telemetry.

The open source package manager that aggregates 34+ skill sources, scans 400K+ skills from registries, auto-translates to 46 agent formats, and adds memory, security scanning, and team workflows. Everything runs locally.

$ skillkit install anthropics/skills

→ Fetched from github.com/anthropics/skills
→ Translated to 46 agent formats
→ Scanned: 0 issues found

$ skillkit install skills.sh/vercel/a

At a Glance

What you get vs. a simple installer.

Capability Simple Installers SkillKit
Agent Support 1-17 agents 46 agents
Sources Single registry 34+ aggregated
Translation None or limited All 45 formats
Memory None Persistent learning
Security None 46-rule scanner
Team Sync None .skills manifest
Telemetry Varies Zero. Ever.
API Access None REST + MCP + Python

Smart Recommendations

AI analyzes your codebase and recommends from 400K+ skills.

Auto Translation

The npm for agent skills. Write once, auto-translate to 45 formats.

Session Memory

AI learnings persist across sessions and projects.

Primer

Auto-generate agent instructions for all 46 agents from your codebase.

Skill Tree

Browse 400K+ skills in a hierarchical taxonomy with 12 categories.

Workflows

Compose multi-step automated skill sequences.

Security Scanner

Detect prompt injection, secrets, and malicious patterns in skills.

Testing

Built-in test framework with assertions.

CI/CD

GitHub Actions, GitLab CI, pre-commit.

Find Skills

Search 400K+ skills across registries

Search
reacttypescriptnextjstestingdockerkubernetespythonsecurity

Submit Your Skill

Share your AI agent skills with the community.
Your GitHub repository URL
Run this command:

$ npx skillkit@latest publish https://github.com/owner/repo

Trending Skills

Popular skills used by the community.

  1. Frontend Design - anthropics/skills - designui
  2. TypeScript Best Practices - anthropics/skills - typescriptjavascript
  3. Test Driven Development - obra/superpowers - testing
  4. Vercel React Best Practices - vercel-labs/agent-skills - reactvercel
  5. Semgrep - trailofbits/skills - typescriptai
  6. Building Native UI - expo/skills - exporeact-native
  7. Senior Architect - sickn33/antigravity-awesome-skills
  8. Supabase Postgres Best Practices - supabase/agent-skills - supabasepostgres

Browse by Category

Official

Compatibility Matrix

Skill support across all 46 agents and 10 categories.

Category ⟁Claude ◫Cursor ◎Codex ✦Gemini ⬡OpenCod ◇Windsur ⌘Copilot ●Cline ◉Devin ◆Roo △Kilo ▽Kiro CL ◊Goose ■Trae ⚡AMP ◦Aider ◐Cody ▣Amazon ↑Antigra
⇄Skill Translation ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ○ ✓ ✓ ✓ ✓ ✓ ✓ ✓
⚓Hooks ✓ ○ ○ ○ ○ ✕ ✕ ✓ ○ ✓ ○ ○ ○ ○ ○ ○ ○ ○ ○
⚙MCP & Tools ✓ ✕ ✕ ✕ ✕ ✕ ✕ ✓ ○ ✓ ✕ ✕ ✕ ✕ ✕ ✕ ✕ ✕ ✕
▣Context Window ✓ ○ ○ ✓ ✓ ○ ○ ✓ ✓ ✓ ✓ ✓ ✓ ○ ○ ○ ○ ○ ○
◐Security Skills ✓ ✓ ○ ✓ ✓ ✓ ○ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ○
◒Testing Skills ✓ ✓ ○ ✓ ✓ ✓ ○ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ○ ○
⚛Frontend Skills ✓ ✓ ○ ✓ ✓ ✓ ○ ✓ ✓ ✓ ○ ○ ○ ✓ ✓ ○ ○ ○ ○
⎓Backend Skills ✓ ✓ ○ ✓ ✓ ✓ ○ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ○ ○
◑DevOps Skills ✓ ✓ ○ ✓ ✓ ○ ○ ✓ ✓ ✓ ○ ○ ○ ○ ○ ○ ○ ○ ○
□Documentation ✓ ✓ ○ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓
Score 100%
20/20
80%
16/20
50%
10/20
85%
17/20
85%
17/20
70%
14/20
50%
10/20
100%
20/20
90%
18/20
100%
20/20
75%
15/20
70%
14/20
75%
15/20
75%
15/20
80%
16/20
70%
14/20
70%
14/20
55%
11/20
50%
10/20

Advanced Capabilities

SkillKit isn't just a package manager. It's a complete infrastructure for intelligent, distributed AI agent orchestration.

AI Generate

Create Skills from Natural Language Generate skills from plain English with multi-source context. Pulls from documentation (Context7), your codebase patterns, 400K+ marketplace skills, and your corrections. Works with any LLM: Claude, GPT-4, Gemini, Ollama, or OpenRouter.

$ skillkit generate

→ Interactive skill wizard

$ skillkit generate --provider openai

→ Use specific LLM

$ skillkit generate --compose "testing"

→ Compose from skills

$ skillkit generate --agents claude,cursor

→ Multi-agent output

Built for Real Workflows

Whether you're a solo developer or managing enterprise infrastructure, SkillKit adapts to your needs.

Multi-Agent Developer

PAIN: You've built skills for Claude Code. Now your team wants Cursor, and you want to try Windsurf.

FIX: SkillKit translates skills between all 46 agents. Write once, deploy everywhere.

$ skillkit translate --to cursor,windsurf

Team Lead

PAIN: Each developer has different skills. Onboarding takes weeks. Knowledge is scattered.

FIX: Shared .skills manifest ensures everyone has the same capabilities. Git-based sync keeps teams aligned.

$ skillkit team init && skillkit team share

New Project Starter

PAIN: Configuring AI agents for each new project is tedious. You forget best practices.

FIX: Primer analyzes your codebase and auto-generates optimized instructions for all agents.

$ skillkit primer --all-agents

Enterprise Architect

PAIN: Running AI agents across multiple machines and keeping them in sync is a nightmare.

FIX: Mesh network with encrypted P2P, peer trust, and inter-agent messaging.

$ skillkit mesh init && skillkit mesh discover

Developer Tools Built-In

  • Workflow Orchestration: Compose skills into multi-step workflows. Chain operations, handle errors, and automate complex tasks.
  • Skill Testing: Built-in test framework with assertions. Validate skills work correctly before deployment.
  • CI/CD Integration: Generate GitHub Actions, GitLab CI, and pre-commit hooks for automated skill management.
  • Slash Commands: Auto-generate /commands for Claude, Cursor, and other agents from your skills.

Skill Sources

35 sources aggregated from open source repositories and registries.