Awesome Agent Skills
概览
作者提供的说明(英文原文)
MCP server with 100+ AI agent skills from Anthropic, Vercel, Trail of Bits, Hugging Face & more.
PopNx MCP 的三大支柱
这三项永远不会合并成一个分数:意见、测量和人气是不同的东西。
- 用户评价i暂无评价评价功能将在后续阶段与账户功能一起开放
- 我们的基础检查i1 项中通过 1 项依据公开信息检查;尚未扫描代码
- 使用情况i非常新↓ 81/周 · ★ –
它能帮你做什么
我们还没有为它写通俗的介绍。下面是我们从注册中心读取到的事实。
- 连接方式: 在你的电脑上运行
- 传输方式: stdio
- 分类: 搜索与 AI 记忆
安装
选择你使用的 AI 应用,然后复制设置。
安装之前
- 基础检查:1 项中通过 1 项
- 在你的电脑上运行,拥有你的用户账户的权限(文件、网络、应用)。
- 需要你自己填写:
SKILLS_REPO_URL,SKILLS_CACHE_DIR,SKILLS_SYNC_INTERVAL,LOG_LEVEL
你需要
- npm已安装 Node.js(用于 npx 命令)
你用的是哪个应用?
在上方选择你的应用。我们会记住它,下次按钮会显示“安装到 …”。
打开 Claude Desktop → Settings → Developer → Edit Config,把下面的内容粘贴到 claude_desktop_config.json
{
"mcpServers": {
"awesome-agent-skills": {
"command": "npx",
"args": [
"-y",
"awesome-agent-skills-mcp"
],
"env": {
"SKILLS_REPO_URL": "your-value",
"SKILLS_CACHE_DIR": "your-value",
"SKILLS_SYNC_INTERVAL": "your-value",
"LOG_LEVEL": "your-value"
}
}
}
}在终端中运行这条命令
claude mcp add awesome-agent-skills --env SKILLS_REPO_URL=your-value --env SKILLS_CACHE_DIR=your-value --env SKILLS_SYNC_INTERVAL=your-value --env LOG_LEVEL=your-value -- npx -y awesome-agent-skills-mcp运行前请先填写空着的值。
或者自己复制设置
粘贴到项目中的 .cursor/mcp.json,或 ~/.cursor/mcp.json
{
"mcpServers": {
"awesome-agent-skills": {
"command": "npx",
"args": [
"-y",
"awesome-agent-skills-mcp"
],
"env": {
"SKILLS_REPO_URL": "your-value",
"SKILLS_CACHE_DIR": "your-value",
"SKILLS_SYNC_INTERVAL": "your-value",
"LOG_LEVEL": "your-value"
}
}
}
}浏览器会先询问是否打开应用,然后应用会显示设置并请你确认。密钥的值会留空,由你自己填写。
或者自己复制设置
粘贴到项目中的 .vscode/mcp.json
{
"servers": {
"awesome-agent-skills": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"awesome-agent-skills-mcp"
],
"env": {
"SKILLS_REPO_URL": "your-value",
"SKILLS_CACHE_DIR": "your-value",
"SKILLS_SYNC_INTERVAL": "your-value",
"LOG_LEVEL": "your-value"
}
}
}
}浏览器会先询问是否打开应用,然后应用会显示设置并请你确认。密钥的值会留空,由你自己填写。
这个服务器在你的电脑上运行,而 ChatGPT 只能连接互联网上的服务器。请选择其他应用。
你的选择通过 Cookie 保存在这台设备上,无需账户。
需要你填写的值
SKILLS_REPO_URLenvURL of the skills repository to sync fromSKILLS_CACHE_DIRenvDirectory path for caching skills dataSKILLS_SYNC_INTERVALenvAuto-sync interval in minutes (0 to disable)LOG_LEVELenvLogging level (debug, info, warn, error)
这些设置是根据作者在注册中心的条目生成的。如果 README 的写法不同,请以 README 为准。
它能做什么
它在你的电脑上运行;安装后,工具列表会显示在你的 AI 应用中。
基础检查i
我们检查的是任何人都能看到的信息;我们没有阅读代码。不适用的检查项不计入总数。 · 检查于 2026年11月8日
- ✓ 通过
作者与仓库一致
注册名称已通过 GitHub 账户 shadowrootdev 验证,该账户拥有此仓库
- … 尚未检查
仍在维护
我们还没有从 GitHub 读取这个仓库
- … 尚未检查
有许可证
我们还没有从 GitHub 读取这个仓库
- – 不适用
端点有响应
它在你的电脑上运行,没有互联网端点
- – 不适用
使用 OAuth 登录
在你的电脑上运行,改用本地设置
- … 尚未检查
工具描述中没有隐藏指令
它在你的电脑上运行;我们只能读取互联网端点的工具列表
与“搜索与 AI 记忆”中的其他服务器比较i
- 星标数多于此分类中其他 5686 个服务器的 0%
- 发布到注册中心的时间比此分类中其他 5686 个服务器的 2% 更新
- 基础检查结果等于或优于此分类中其他 5686 个服务器的 100%
与注册中心里这个分类中所有活跃的服务器比较。只有当发布者拥有该仓库时,星标数才计入。
常见问题
Awesome Agent Skills 安全吗?
它在 1 项适用的基础检查中通过了 1 项。我们没有阅读代码。在添加密钥或令牌之前,请先看看这个服务器要求哪些权限,并使用只具备所需权限的密钥。
它需要哪些密钥或代码?
注册中心没有列出必需的密钥,不过有些服务器会在首次使用时要求你登录。
它在我的电脑上运行还是在云端?
它在你的电脑上运行。你的 AI 应用会在需要时启动它,数据不会经过作者的服务器(除非它自己调用在线服务)。
它还在维护吗?
GitHub 仓库最后更新于–,最新版本(1.0.2)于9个月前进入注册中心。
哪些应用可以使用它?
支持 MCP 的应用,例如 Claude Desktop、Claude Code、Cursor 和 VS Code。你的电脑上还需要有 Node.js。
有什么可以替代它?
同一分类中有:Context7, Codebase Memory, Exa
替代选择
同一分类中其他作者的服务器,按星标数排序。
如何看卡片:↓ 每周下载量i★ GitHub 星标i三大支柱:评价 · 检查 · 使用情况i更新 = 最新版本发布到注册中心的时间i
Codebase Memory
Codebase knowledge graph for AI agents — 162 languages, sub-ms queries, 99% fewer tokens.
Exa
Connect AI agents to Exa for web search, content fetching, and multi-step research.
作者的 README
打开 README
按原文从作者的仓库复制。PopNx MCP 没有审查或背书其内容。HTML、图片和脚本已被移除。链接会离开本站。
Awesome Agent Skills MCP Server
A Model Context Protocol (MCP) server that provides access to 100+ curated AI agent skills from the [VoltAgent Awesome Agent Skills](https://github.com/VoltAgent/awesome-agent-skills) collection.
Transform your AI assistants into specialized experts with skills from Anthropic, Vercel, Trail of Bits, Hugging Face, Stripe, Expo, and many more leading organizations.
🚀 Quick Start
npx awesome-agent-skills-mcp
That's it! Add to your MCP client config and start using 100+ AI skills.
Table of Contents
- Features
- Quick Start
- Installation
- Configuration
- VS Code / GitHub Copilot
- Claude Desktop
- OpenCode
- Available Skills
- MCP Tools
- Usage Examples
- Development
- API Reference
- Troubleshooting
- Credits
- License
Features
- 100+ Curated Skills - Access skills from top organizations including Anthropic, Vercel, Trail of Bits, Hugging Face, and more
- Auto-Sync - Automatically fetches and updates skills from the VoltAgent repository
- MCP 2024-11-05 Compliant - Full compatibility with the latest Model Context Protocol specification
- Multi-Client Support - Works with Claude, GitHub Copilot, OpenCode, and any MCP-compatible client
- Smart Caching - Efficient JSON-based caching for fast startup times
- Type-Safe - Built with TypeScript and Zod for runtime validation
- Zero Configuration - Works out of the box with sensible defaults
Quick Start
Using npx (Recommended)
npx awesome-agent-skills-mcp
Global Installation
npm install -g awesome-agent-skills-mcp
awesome-agent-skills-mcp
Installation
Prerequisites
- Node.js >= 20.0.0
- npm or yarn
From npm
npm install awesome-agent-skills-mcp
From Source
git clone https://github.com/shadowrootdev/awesome-agent-skills-mcp.git
cd awesome-agent-skills-mcp npm install npm run build
Configuration
VS Code / GitHub Copilot
Create or update .vscode/mcp.json in your project:
{
"servers": { "awesome-agent-skills": { "command": "npx", "args": ["awesome-agent-skills-mcp"] } } }
Or with a local installation:
{
"servers": { "awesome-agent-skills": { "command": "node", "args": ["/path/to/awesome-agent-skills-mcp/dist/index.js"] } } }
Important: After adding the configuration, fully quit VS Code (Cmd+Q / Alt+F4) and reopen it for changes to take effect.
Claude Desktop
Add to your Claude Desktop configuration (~/Library/Application Support/Claude/claude_desktop_config.json on macOS):
{
"mcpServers": { "awesome-agent-skills": { "command": "npx", "args": ["awesome-agent-skills-mcp"] } } }
OpenCode
Add to ~/.config/opencode/opencode.json:
{
"mcp": { "awesome-agent-skills": { "type": "local", "command": ["npx", "awesome-agent-skills-mcp"], "enabled": true } } }
Environment Variables
| Variable | Default | Description |
|---|---|---|
SKILLS_REPO_URL | https://github.com/VoltAgent/awesome-agent-skills.git | Skills repository URL |
SKILLS_CACHE_DIR | .cache | Cache directory path |
SKILLS_SYNC_INTERVAL | 60 | Auto-sync interval in minutes (0 to disable) |
LOG_LEVEL | info | Logging level (debug, info, warn, error) |
Available Skills
The server provides access to 100+ skills from leading organizations:
Anthropic
Document processing, presentation creation, spreadsheet manipulation, PDF handling, algorithmic art, MCP building, and more.
| Skill | Description |
|---|---|
docx | Create, edit, and analyze Word documents |
pptx | PowerPoint presentation creation and editing |
xlsx | Spreadsheet manipulation with formulas |
pdf | PDF processing and form filling |
mcp-builder | Guide for creating MCP servers |
webapp-testing | Playwright-based web app testing |
Vercel
React and Next.js best practices, deployment, and performance optimization.
| Skill | Description |
|---|---|
react-best-practices | React performance optimization guidelines |
next-best-practices | Next.js conventions and patterns |
web-design-guidelines | UI/UX compliance review |
vercel-deploy | Deploy apps to Vercel |
Trail of Bits
Security analysis, smart contract auditing, and code review tools.
| Skill | Description |
|---|---|
building-secure-contracts | Smart contract security toolkit |
semgrep-rule-creator | Create custom Semgrep rules |
property-based-testing | Property-based testing guidance |
static-analysis | Static analysis tooling |
Hugging Face
ML model training, dataset management, and Hub operations.
| Skill | Description |
|---|---|
hugging-face-cli | HF Hub CLI operations |
hugging-face-datasets | Dataset creation and management |
hugging-face-model-trainer | Model fine-tuning with TRL |
hugging-face-evaluation | Model evaluation workflows |
Sentry
Code review, commit conventions, and PR automation.
| Skill | Description |
|---|---|
code-review | Sentry engineering code review practices |
commit | Conventional commit messages |
create-pr | PR creation following Sentry conventions |
find-bugs | Bug and vulnerability detection |
And Many More...
- Stripe - Payment integration best practices
- Expo - React Native app development
- n8n - Workflow automation patterns
- Sanity - CMS best practices
- Neon - Serverless Postgres
- Remotion - Programmatic video creation
MCP Tools
The server exposes four MCP tools:
list_skills
List all available skills with optional filtering.
// List all skills
{ }
// Filter by source { "source": "repository" }
// Filter by tag { "tag": "security" }
get_skill
Get detailed information about a specific skill.
{ "skill_id": "react-best-practices" }
invoke_skill
Invoke a skill with optional parameters.
{
"skill_id": "docx", "parameters": { "document_type": "report" } }
refresh_skills
Manually trigger a skills refresh from the repository.
{ }
Usage Examples
In GitHub Copilot Chat
@workspace Use the react-best-practices skill to review my React components
@workspace List all available security-related skills
@workspace Get the stripe-best-practices skill and apply it to my checkout code
In Claude
What skills are available for Next.js development?
Use the code-review skill to analyze my pull request
Development
Setup
git clone https://github.com/shadowrootdev/awesome-agent-skills-mcp.git
cd awesome-agent-skills-mcp npm install
Build
npm run build
Run Tests
# Run all tests
npm test
Run tests in watch mode
npm run test:watch
Run integration tests only
npm run test:integration
Lint & Format
npm run lint
npm run format
Project Structure
awesome-agent-skills-mcp/
├── src/ │ ├── index.ts # Entry point │ ├── server.ts # MCP server implementation │ ├── config.ts # Configuration management │ ├── models/ │ │ ├── skill.ts # Skill type definitions │ │ ├── parameter.ts # Parameter schemas │ │ ├── registry.ts # SkillRegistry class │ │ └── repository.ts # Repository source model │ ├── services/ │ │ ├── git-sync.ts # Git repository sync │ │ ├── skill-parser.ts # Skill parsing from README │ │ └── skill-executor.ts # Skill invocation │ ├── cache/ │ │ └── cache-manager.ts # JSON-based caching │ └── utils/ │ └── logger.ts # Structured logging ├── tests/ │ ├── unit/ # Unit tests │ └── integration/ # Integration tests ├── dist/ # Compiled output └── .cache/ # Runtime cache (gitignored)
API Reference
Skill Object
interface Skill {
id: string; // Unique identifier name: string; // Display name description: string; // Short description source: 'repository' | 'local'; sourcePath: string; // GitHub URL or local path content: string; // Full skill content (markdown) parameters: ParameterSchema[]; metadata: { author?: string; version?: string; tags?: string[]; requirements?: string[]; sourceOrg?: string; // GitHub organization sourceRepo?: string; // GitHub repository }; lastUpdated: Date; }
Parameter Schema
interface ParameterSchema {
name: string; type: 'string' | 'number' | 'boolean' | 'object' | 'array'; description: string; required: boolean; default?: unknown; enum?: unknown[]; }
Troubleshooting
Skills Not Loading
- Check cache: Delete
.cachedirectory and restart - Verify network: Ensure access to GitHub
- Check logs: Set
LOG_LEVEL=debugfor verbose output
GitHub Copilot Not Seeing Skills
- Full restart required: Quit VS Code completely (Cmd+Q) and reopen
- Check config path: Ensure
.vscode/mcp.jsonis in the workspace root - Verify server: Run
npx awesome-agent-skills-mcpmanually to test
Connection Errors
# Test the server manually
echo '{"jsonrpc":"2.0","id":1,"method":"initialize","params":{"protocolVersion":"2024-11-05","capabilities":{},"clientInfo":{"name":"test","version":"1.0.0"}}}' | npx awesome-agent-skills-mcp
Credits
This project is built on top of the incredible work by the open-source community:
Data Source
[VoltAgent Awesome Agent Skills](https://github.com/VoltAgent/awesome-agent-skills) - A curated collection of AI agent skills maintained by VoltAgent. This MCP server fetches and serves skills from this repository, making them accessible to MCP-compatible AI assistants.
Skill Contributors
Skills in this
只显示了开头部分;其余内容请在仓库中阅读