🍺 Bars MCP Server

Bars is a comprehensive library documentation service that provides up-to-date documentation and code examples for any library. It works as a Model Context Protocol (MCP) server for AI assistants like Cursor, Claude, VS Code, and more.

✨ Features

• 🔍 Smart Search: Find libraries quickly with intelligent search
• 📚 Rich Documentation: Get comprehensive documentation with code examples
• ⚡ Fast & Reliable: Powered by Firebase with sub-second response times
• 🔧 MCP Compatible: Works with all major AI assistants
• 🌐 Live API: Access to live documentation service

🚀 Quick Start

Install in Cursor

Go to: Settings -> Cursor Settings -> MCP -> Add new global MCP server

{
  "mcpServers": {
    "bars": {
      "command": "npx",
      "args": ["-y", "bars-mcp-server"]
    }
  }
}

Install in Claude Desktop

Add to your claude_desktop_config.json:

{
  "mcpServers": {
    "bars": {
      "command": "npx",
      "args": ["-y", "bars-mcp-server"]
    }
  }
}

Install in VS Code

Add to your VS Code MCP config:

{
  "servers": {
    "Bars": {
      "type": "stdio",
      "command": "npx",
      "args": ["-y", "bars-mcp-server"]
    }
  }
}

Install in Zed

Add to your Zed settings.json:

{
  "context_servers": {
    "Bars": {
      "command": {
        "path": "npx",
        "args": ["-y", "bars-mcp-server"]
      },
      "settings": {}
    }
  }
}

Install in Windsurf

Add to your Windsurf MCP config:

{
  "mcpServers": {
    "bars": {
      "command": "npx",
      "args": ["-y", "bars-mcp-server"]
    }
  }
}

🛠️ Usage

Once installed, you can use Bars in your AI assistant:

Search for React documentation using Bars

Get Next.js routing examples with Bars

Find TypeScript best practices using Bars

🔧 Available Tools

• resolve-library-id: Search for libraries and get Bars-compatible IDs
• get-library-docs: Fetch comprehensive documentation for any library

🌐 Live Service

• Website: https://bars-fca85.web.app
• API: https://api-h76uo622iq-uc.a.run.app

⚙️ Configuration

Environment Variables

{
  "mcpServers": {
    "bars": {
      "command": "npx",
      "args": ["-y", "bars-mcp-server"],
      "env": {
        "DEFAULT_MINIMUM_TOKENS": "10000"
      }
    }
  }
}

Alternative Runtimes

Using Bun

{
  "mcpServers": {
    "bars": {
      "command": "bunx",
      "args": ["-y", "bars-mcp-server"]
    }
  }
}

Using Deno

{
  "mcpServers": {
    "bars": {
      "command": "deno",
      "args": ["run", "--allow-env", "--allow-net", "npm:bars-mcp-server"]
    }
  }
}

🐳 Docker Support

FROM node:20-alpine
RUN npm install -g bars-mcp-server
CMD ["bars-mcp"]

🧪 Testing

Test the installation:

echo '{"jsonrpc": "2.0", "id": 1, "method": "tools/call", "params": {"name": "resolve-library-id", "arguments": {"libraryName": "react"}}}' | npx -y bars-mcp-server

🔧 Development

Local Development

git clone https://github.com/doctorai/bars-mcp-server.git
cd bars-mcp-server
npm install
npm run build
npm test

Building

npm run build

📝 API Reference

Search Libraries

GET https://api-h76uo622iq-uc.a.run.app/v1/search?query={library_name}

Get Documentation

GET https://api-h76uo622iq-uc.a.run.app/v1/{library_path}?tokens={count}&topic={topic}

🤝 Contributing

1. Fork the repository
2. Create a feature branch
3. Make your changes
4. Test thoroughly
5. Submit a pull request

📄 License

MIT License - see LICENSE file for details.

🆚 Comparison with Context7

Bars provides the same functionality as Context7 but with:

• ✅ Your own infrastructure
• ✅ Firebase-powered reliability
• ✅ Custom branding
• ✅ Full control over data and features

📞 Support

• 🌐 Website: https://bars-fca85.web.app
• 🐛 Issues: https://github.com/doctorai/bars-mcp-server/issues
• 📧 Email: [email protected]

Built with ❤️ using Firebase, TypeScript, and the Model Context Protocol.