robot

This MCP server is designed to control a robot via the Model Context Protocol (MCP). It exposes a FastMCP-based interface that enables actions such as moving the robot and responding to MCP requests from clients. The project is scaffolded to run in both stdio mode and an HTTP API mode, allowing you to interact with the robot either through the local process communication channel or over HTTP. You can use the provided npm scripts to start the server in the mode you prefer and then send MCP commands to trigger actions or query state.

In standard usage, install dependencies, start the stdio server, and optionally start the HTTP server for remote access. The templates typically provide a set of MCP actions (e.g., move, rotate) and a way to subscribe to contextual data from the robot. If you’re developing or extending the server, you can run development builds to see live changes and test MCP messages in real time.

Prerequisites:

• Node.js (LTS version)
• npm (comes with Node.js)

Installation steps:

1. Clone the repository or download the project archive

   • git clone https://github.com/your-organization/robot-mcp-server.git

2. Navigate into the project directory

   • cd robot-mcp-server

3. Install dependencies

   • npm install

4. Start the server (stdio)

   • npm start

5. (Optional) Start the HTTP server for remote access

   • npm run start:http

Development workflow (optional):

• For development with stdio: npm run dev
• For development with HTTP: npm run dev:http

Notes:

• Ensure any robot hardware interfaces are properly connected and configured as expected by the server.
• If port or host configurations are required, adjust environment variables or configuration files as documented in the project.

Tips and common issues:

• If the HTTP server fails to start, verify that the port is not in use by another process and that network permissions allow binding to the chosen port.
• Check that any required robot drivers or SDKs are installed and accessible in your environment.
• If you modify MCP actions, ensure the actions are correctly registered with the MCP framework and that input payloads conform to expected schemas.
• Environment variables you may encounter include PORT for the HTTP server, DEBUG for verbose logging, and any hardware-specific configuration (e.g., ROBOT_SERVO_PORT).
• For troubleshooting, run in development mode to get live reloads and clearer error messages.