A Model Context Protocol (MCP) server for Uptime Kuma v2 🐻🤖

This MCP server enables AI assistants to interact with Uptime Kuma, allowing them to manage monitoring services, add/update/remove monitors, and control monitor states.

• 🔐 Authentication: Support for both username/password and API key authentication
• 📊 Monitor Management: Add, update, and remove monitors
• ⏯️ Monitor Control: Pause and resume monitors
• 📋 Monitor Information: Get details of specific monitors or list all monitors
• 🐳 Docker Support: Includes docker-compose.yml for easy Uptime Kuma deployment
• 🏗️ TypeScript: Fully typed with Zod schema validation
• ⚡ Built with Bun: Fast runtime and package manager (requires Node.js v18+ LTS)

• Bun installed on your system
• Node.js v18+ LTS
• An Uptime Kuma v2 instance (can be started with included docker-compose.yml)

bun install

bun run build

If you don't have an Uptime Kuma instance running, use the included docker-compose:

docker-compose up -d

Uptime Kuma will be available at http://localhost:3001. Go to the URL and finish the uptime-kuma configuration.

Set the credentials in .env:

cp .env.example .env

bun run dev

Or use the built version:

bun run build
bun dist/index.js

The MCP Inspector is a visual tool for testing and debugging MCP servers. It provides an interactive interface to explore your server's capabilities, test tools, and inspect responses.

Start your MCP server with the Inspector:

bunx @modelcontextprotocol/inspector bun run dev

or the built version

bun run build
bunx @modelcontextprotocol/inspector bun run dist/index.js

The Inspector will:

1. Start your MCP server
2. Open a web interface (typically at http://localhost:6274)
3. Allow you to interactively test all available tools
4. Display real-time request/response data
5. Help debug issues with your server

Once the Inspector is running:

1. Connect to Server: The Inspector automatically connects to your MCP server
2. Browse Tools: View all available tools (add_monitor, list_monitors, etc)
3. Test Tools: Fill in parameters and execute tools to see responses
4. Inspect Results: View detailed request/response JSON
5. Debug Issues: Check for errors and validation problems

This is especially useful for:

• Testing monitor creation with different configurations
• Verifying authentication is working correctly
• Exploring the response format of each tool
• Debugging issues before integrating with Claude Desktop or other MCP clients

To use this server with an MCP client like Claude Desktop, add it to your MCP configuration:

Add to your claude_desktop_config.json:

{
  "mcpServers": {
    "kuma-mcp": {
      "command": "node",
      "args": ["/path/to/kuma-mcp/dist/index.js"],
      "env": {
        "UPTIME_KUMA_URL": "http://localhost:3001",
        "UPTIME_KUMA_USERNAME": "your-username",
        "UPTIME_KUMA_PASSWORD": "your-password"
        // Alternative: Use API key instead of username/password
        // "UPTIME_KUMA_API_KEY": "your-api-key"
      }
    }
  }
}

• add_monitor
• update_monitor
• remove_monitor
• pause_monitor
• resume_monitor
• get_monitor
• list_monitors

For more information about the Uptime Kuma API, see:

• Uptime Kuma API Documentation
• uptime-kuma-api - A Python wrapper for the Uptime Kuma Socket.IO API

Contributions are welcome! Please feel free to submit a Pull Request.

This project uses Semantic Commit Messages. See CONTRIBUTING.md for detailed guidelines.

MIT

• Model Context Protocol
• Uptime Kuma
• Uptime Kuma API Documentation