Sandbox
@rohitg00/skillkit

Skill manager for Claude Code, Cursor, and Codex

SkillKit is a CLI for working with reusable agent skills across many editors and coding assistants. It installs skills from local paths, Git repos, gists, and a marketplace, then translates them into each agent’s expected format and folder. It also adds commands for sync, recommendation, memory, publishing, and a REST or MCP setup.

1,495 stars142 forksTypeScriptUpdated 3mo ago
Who it's for

Builders who want to manage one skill set across Claude Code, Cursor, Codex, Copilot, and other agent tools.

What it delivers

You can install a skill once and ship it to many agents without rewriting it by hand.

What it does

Install skills from many sources

Add skills from GitHub, GitLab, local paths, gists, or the marketplace with `skillkit add`.

Translate skills between agent formats

Convert one skill into formats like `SKILL.md` or `.mdc` with `skillkit translate`.

Sync skills into agent folders

Write installed skills into the right config directories with `skillkit sync`.

Recommend skills from your stack

Inspect a repo and suggest relevant skills with `skillkit recommend`.

Generate instructions and skills

Create agent instruction files and draft skills from plain English with `skillkit primer` and `skillkit generate`.

Support memory and collaboration

Capture session learnings, export them, and share skills through a committable manifest.

Expose runtime access

Run a REST server, MCP server, TUI, and mesh commands from the optional packages.

How to get it

  1. 1Run
    npx skillkit init                     # detect agent, create dirs
    skillkit recommend                    # stack-aware suggestions
    skillkit add anthropics/skills        # install from marketplace
    skillkit sync                         # deploy to agent config
  2. 2Pick one. All three do the same thing.
    npm install -g skillkit       # npm
    pnpm add -g skillkit          # pnpm
    bun add -g skillkit           # bun
  3. 3Skipped optional? Add just the one you want later
    npm i -g @skillkit/tui         # enables: skillkit ui
    npm i -g @skillkit/api         # enables: skillkit serve
    npm i -g @skillkit/mesh        # enables: skillkit mesh
    npm i -g @skillkit/messaging   # enables: skillkit message
  4. 4npx skillkit add works with zero install. First call caches the package under…
    npx skillkit add anthropics/skills                    # full
    npx --omit=optional skillkit add anthropics/skills    # slim

README

SkillKit. One skill. Every agent.SkillKit. One skill. Every agent.

One skill. Every agent. 46 of them.


npm downloads stars CI license

400K+ skills 46 agents 31 sources 757 tests Apache 2.0 PRs welcome

slim install v1.24.0 new one skill every agent

Quick nav

Quick start · Install · Commands · Agents · Sources · API · Website


The problem

Every AI coding agent wants skills. Every agent invented a different format.

AgentFormatDirectory
Claude CodeSKILL.md.claude/skills/
Cursor.mdc.cursor/skills/
CopilotMarkdown.github/skills/
WindsurfMarkdown.windsurf/skills/
... 41 more

You rewrite the same skill for each agent. Or you lock in to one.

The fix

SkillKit is the package manager for AI agent skills. Install from 400K+ skills across 31 sources. Auto-translate between formats. Persist session learnings. Ship to 46 agents at once.

npx skillkit add anthropics/skills

That is the whole first-run. Pick your agent (your detected agent is pre-selected), confirm, done.

Quick start

npx skillkit init                     # detect agent, create dirs
skillkit recommend                    # stack-aware suggestions
skillkit add anthropics/skills        # install from marketplace
skillkit sync                         # deploy to agent config

Four commands. Your agent now has PDF processing, code review, auth patterns, and whatever you need.

What the first install looks like

$ npx skillkit add anthropics/skills

 ◇ Detected 32 agents
 │
 ◆ Install to
 │  ● Just Claude Code (detected)      claude-code
 │  ○ Select specific agents           space to toggle
 │  ○ All supported agents             32 agents, writes to every adapter
 └

 ◇ Cloning anthropics/skills
 ◇ Security scan: 42/42 skills pass
 ◇ Installed 42 skills to Claude Code
 │
 └ Done in 3.1s. Run `skillkit list` to see them.

Install

Pick one. All three do the same thing.

npm install -g skillkit       # npm
pnpm add -g skillkit          # pnpm
bun add -g skillkit           # bun

Both skillkit and sk work as the binary name.

Full vs slim

Four features ship as optional packages so a bare CLI stays small. The default install pulls everything. Add --omit=optional if you only want the core.

FeaturePackageCommand
Terminal UI@skillkit/tuiskillkit ui
REST server@skillkit/apiskillkit serve
Peer mesh@skillkit/meshskillkit mesh
Agent messaging@skillkit/messagingskillkit message

Cold install numbers on a fresh npm cache:

118 packages 9s install 0 vulnerabilities

ModePackagesTimeVulns (crit/high)
npm i -g skillkit41818s0/0
npm i -g skillkit --omit=optional1189s0/0

Skipped optional? Add just the one you want later:

npm i -g @skillkit/tui         # enables: skillkit ui
npm i -g @skillkit/api         # enables: skillkit serve
npm i -g @skillkit/mesh        # enables: skillkit mesh
npm i -g @skillkit/messaging   # enables: skillkit message

The CLI catches missing optional packages and prints a one-line hint. No stack trace.

Using npx

npx skillkit add <owner/repo> works with zero install. First call caches the package under ~/.npm/_npx/. Every run after that is instant until a new version ships.

npx skillkit add anthropics/skills                    # full
npx --omit=optional skillkit add anthropics/skills    # slim

If you reach for npx skillkit more than twice, switch to a global install. Kills the prompt-to-proceed and the per-release refetch.

What you can do

Install from anywhere
skillkit add anthropics/skills           # GitHub
skillkit add gitlab:team/skills          # GitLab
skillkit add ./my-local-skills           # local path
skillkit add https://gist.github.com/... # gist
Translate between agents

Write once for Claude, ship everywhere:

skillkit translate my-skill --to cursor
skillkit translate --all --to windsurf,codex
skillkit translate my-skill --to copilot --dry-run
Stack-aware recommendations

SkillKit reads your repo, spots your stack, ranks skills:

$ skillkit recommend

  92% vercel-react-best-practices
  87% tailwind-v4-patterns
  85% nextjs-app-router
  81% shadcn-ui-components
Runtime skill discovery (REST)

Start the server, let agents fetch skills on demand:

skillkit serve
# http://localhost:3737

curl "http://localhost:3737/search?q=react+performance"

Or wire it up with MCP:

{
  "mcpServers": {
    "skillkit": { "command": "npx", "args": ["@skillkit/mcp"] }
  }
}

Or call from Python:

from skillkit import SkillKitClient

async with SkillKitClient() as client:
    results = await client.search("react performance", limit=5)

REST docs · MCP docs · Python client

Auto-generate agent instructions

Analyze the codebase and write CLAUDE.md, .cursorrules, AGENTS.md, and friends:

skillkit primer --all-agents
Session memory

AI agents learn during a session, then forget. SkillKit captures what they learned:

skillkit memory compress
skillkit memory search "auth patterns"
skillkit memory export auth-patterns
AI skill generation

Generate skills from plain English with multi-source context:

skillkit generate

Pulls context from four places: Context7 docs, your codebase, 400K marketplace skills, your memory. Works with Claude, GPT-4, Gemini, Ollama (local), or any OpenRouter model.

Team collaboration

Share skills via a committable .skills manifest:

skillkit manifest init
skillkit manifest add anthropics/skills
git commit -m "add team skills"

Everyone else runs skillkit manifest install and matches state.

Mesh network

Distribute agents across machines with encrypted P2P:

skillkit mesh init
skillkit mesh discover
Chrome extension

Save any webpage as a skill from the browser. Click the extension, the page round-trips through the SkillKit API (Turndown, 5-source tag detection, GitHub URL support), the SKILL.md downloads, then run skillkit add ~/Downloads/skillkit-skills/<name>.

Extension docs

Interactive TUI
skillkit ui

h home · m marketplace · r recommend · t translate · i installed · s sync · q quit

SkillKit interactive TUI

Commands

Core
skillkit add <source>            # install skills (live progress)
skillkit remove <skills>         # remove
skillkit remove --source org/repo # bulk by source
skillkit remove --all            # remove everything
skillkit update                  # update (change detection)
skillkit check                   # updates, quality, activity
skillkit translate <skill> --to  # agent format conversion
skillkit sync                    # deploy to agent config
skillkit recommend               # smart recommendations
skillkit generate                # AI skill wizard
skillkit serve                   # REST API server
skillkit publish submit          # publish to marketplace
Discovery & security
skillkit marketplace             # browse
skillkit tree                    # taxonomy
skillkit find <query>            # quick search
skillkit scan <path>             # security scan
Custom sources
skillkit tap add owner/repo
skillkit tap list
skillkit tap remove owner/repo
Issue planner
skillkit issue plan "#42"
skillkit issue plan owner/repo#42
skillkit issue list
Session intelligence
skillkit timeline                # unified event stream
skillkit session handoff         # agent-to-agent context
skillkit lineage                 # skill impact graph
skillkit session explain         # human summary
skillkit activity                # activity log
Advanced
skillkit primer --all-agents     # agent instruction files
skillkit memory compress         # capture session learnings
skillkit mesh init               # multi-machine distribution
skillkit message send            # inter-agent messaging
skillkit workflow run <name>     # run workflows
skillkit test                    # test skills
skillkit cicd init               # CI/CD templates

Full command reference

Supported agents

Top 11
AgentFormatDirectory
Claude CodeSKILL.md.claude/skills/
Cursor.mdc.cursor/skills/
CodexSKILL.md.codex/skills/
Gemini CLISKILL.md.gemini/skills/
OpenCodeSKILL.md.opencode/skills/
GitHub CopilotMarkdown.github/skills/
WindsurfMarkdown.windsurf/skills/
DevinMarkdown.devin/skills/
AiderSKILL.md.aider/skills/
CodySKILL.md.cody/skills/
Amazon QSKILL.md.amazonq/skills/
Plus 35 more

Amp, Antigravity, Augment Code, Bolt, Clawdbot, Cline, CodeBuddy, CodeGPT, CommandCode, Continue, Crush, Droid, Factory, Goose, Hermes Agent, Kilo Code, Kiro CLI, Lovable, MCPJam, Mux, Neovate, OpenClaw, OpenHands, Pi, PlayCode, Qoder, Qwen, Replit Agent, Roo Code, Tabby, Tabnine, Trae, Vercel, Zencoder, Universal.

Full agent details

Creating skills

skillkit create my-skill

Or author a SKILL.md by hand:

---
name: my-skill
description: What this skill does
license: MIT
---

# My Skill

Instructions for the agent.

## When to use
- scenario 1
- scenario 2

## Steps
1. first
2. second

Programmatic API

import { translateSkill, analyzeProject, RecommendationEngine } from 'skillkit';

const skill = await translateSkill(content, 'cursor');

const profile = await analyzeProject('./my-project');
const engine = new RecommendationEngine();
const recs = await engine.recommend(profile);
import { startServer } from '@skillkit/api';
await startServer({ port: 3737, skills: [...] });
import { MemoryCache, RelevanceRanker } from '@skillkit/core';
const cache = new MemoryCache({ maxSize: 500, ttlMs: 86_400_000 });
const ranker = new RelevanceRanker();
const results = ranker.rank(skills, 'react performance');

Skill sources

All original creators credited. Licenses preserved. PRs welcome.

Official partners

RepositoryWhat
anthropics/skillsOfficial Claude Code skills
vercel-labs/agent-skillsNext.js, React
expo/skillsExpo mobile
remotion-dev/skillsProgrammatic video
supabase/agent-skillsDatabase, auth
stripe/aiPayments

Community

trailofbits/skills · obra/superpowers · wshobson/agents · ComposioHQ/awesome-claude-skills · travisvn/awesome-claude-skills · langgenius/dify · better-auth/skills · onmax/nuxt-skills · elysiajs/skills · kadajett/agent-nestjs-skills · cloudai-x/threejs-skills · dimillian/skills · waynesutton/convexskills · kepano/obsidian-skills · giuseppe-trisciuoglio/developer-kit · openrouterteam/agent-skills

Want your skills listed? Submit your repo.

Contributing

Issues and PRs land fast. See CONTRIBUTING.md for the short version.

License

Apache 2.0. See LICENSE.


Docs · Website · API Explorer · npm · GitHub

Built for agents. Written by humans.

Files in the repo

Repository payload31 top-level entries
  • .github
  • apps
  • assets
  • clients
  • docs
  • e2e
  • marketplace
  • packages
  • registry
  • schemas
  • scripts
  • skills
  • src
  • tests
  • .gitignore
  • .npmignore
  • CONTRIBUTING.md
  • LICENSE
  • package-lock.json
  • package.json
  • pnpm-lock.yaml
  • pnpm-workspace.yaml
  • README.md
  • skillkit.gif
  • skills-lock.json
  • SOUL.md
  • tsconfig.base.json
  • tsconfig.json
  • tsup.config.ts
  • turbo.json
  • vitest.config.ts

Discussion (0)

Ask about usage, or say what you built with it

Sign in to join the discussion.

No comments yet. Be the first to say what this is good for.

More tools

JuliusBrussee/
caveman

🪨 why use many token when few token do trick — Claude Code skill that cuts 65% of tokens by talking like caveman

105k
1 add
MemPalace/
mempalace

The best-benchmarked open-source AI memory system. And it's free.

59k
stablyai/
orca

Orca is the ADE for working with a fleet of parallel agents. Run any coding agent with your own subscription. Available on desktop, mobile and remote runtime.

66k

A cross-platform desktop All-in-One assistant for Claude Code, Codex, OpenCode, OpenClaw, Grok Build & Hermes Agent. Only official website: ccswitch.io

132k

Never stop coding. Free MIT AI gateway: one endpoint, 352 providers (150+ free), 1200+ models Kimi, Claude, GPT, Gemini, GLM, DeepSeek, MiniMax. Works with Claude Code, Codex, Cursor, OpenCode, Cline & Copilot. Quota-aware auto-fallback, RTK+Caveman compression saves 15-95% tokens, MCP/A2A, Desktop/PWA. Built by 550+ contributors

64k
headroomlabs-ai/
headroom

Compress tool outputs, logs, files, and RAG chunks before they reach the LLM. 20% fewer tokens for coding agents, 60-95% fewer tokens for JSON, same answers. Library, proxy, MCP server.

71k