Reqable MCP Server

Reqable MCP Server is a command-line tool that connects your AI assistant to Reqable via the Model Context Protocol (MCP).

It exposes Reqable's APIs as MCP tools, enabling AI models to interact with Reqable's debugging and traffic management features.

Installation

npm install -g reqable-mcp-server

Note: npm automatically downloads only the native binary matching your platform — no need to worry about other platform binaries.

Usage

reqable-mcp-server --help

Options

| Option | Alias | Description | Default | |--------|-------|-------------|---------| | --host | -h | Reqable app host | 127.0.0.1 | | --port | -p | Reqable app port | 9000 | | --scope | -s | Tool scope (minimal, full) | minimal |

Example

reqable-mcp-server --host 127.0.0.1 --port 9000 --scope full

MCP Configuration

Add to your AI assistant's MCP configuration (e.g., Claude Desktop, VS Code Copilot):

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

Or with a local installation:

{
  "mcpServers": {
    "reqable-mcp": {
      "command": "reqable-mcp-server",
      "args": []
    }
  }
}

Supported Platforms

| Platform | Architecture | Status | |----------|-------------|--------| | macOS | ARM64 (Apple Silicon) | ✅ | | macOS | x64 (Intel) | ✅ | | Linux | x64 | ✅ | | Windows | x64 | ✅ |

How It Works

This package uses platform-specific optional dependencies — the same pattern used by esbuild, swc, and other popular tools. When you npm install, npm automatically downloads only the native binary for your platform:

| Platform Package | OS | CPU | |-----------------|----|----| | reqable-mcp-server-darwin-arm64 | macOS | arm64 (Apple Silicon) | | reqable-mcp-server-darwin-x64 | macOS | x64 (Intel) | | reqable-mcp-server-linux-x64 | Linux | x64 | | reqable-mcp-server-win32-x64 | Windows | x64 |

The Node.js entry script (bin/mcp-server.js) locates the binary from the matching platform package and spawns it. This approach provides:

• Performance: Native execution speed via Dart AOT compilation
• Efficient installs: Only ~6MB download (single platform binary), not ~25MB (all platforms)
• Zero dependencies: No Dart SDK or runtime required at install time

License

MIT