Ros_mcp_server

What is ROS MCP Server?

ROS MCP Server connects large language models (LLMs) such as Claude, GPT, and Gemini to robots, enabling bidirectional communication without modifying existing robot source code. It is built on the open MCP standard and works with any MCP-compatible client like Claude Desktop, ChatGPT, Cursor, and Gemini CLI.

How to use ROS MCP Server?

Follow the installation guide to get started. Add a rosbridge node to your existing ROS setup and configure the server to work with Claude Code, Codex CLI, Gemini CLI, Claude Desktop, ChatGPT, Cursor, or any other MCP-compatible client.

Key features of ROS MCP Server

• No robot source code changes required.
• True two-way communication — control and observe robots.
• Full context: publish/subscribe to topics, call services and actions.
• Deep ROS understanding guides LLMs on available topics and types.
• Works with any MCP client.
• Compatible across ROS 2 (Jazzy, Humble) and ROS 1 distros.

Use cases of ROS MCP Server

• AI agent diagnosis and debugging of industrial robot end effectors.
• Natural language control of mobile manipulators (e.g., "Grab a Coke from the fridge").
• Controlling simulated robots (e.g., Unitree Go2 in NVIDIA Isaac Sim).
• Real-time monitoring and parameter adjustment of ROS systems.

FAQ from ROS MCP Server

Do I need to change my existing robot code?

No. You only need to add the rosbridge node to your existing ROS setup.

Which LLM clients are supported?

It works with any MCP-compatible client, including Claude Code, Codex CLI, Gemini CLI, Claude Desktop, ChatGPT, and Cursor.

Which ROS versions are supported?

Both ROS 2 (Jazzy, Humble, and others) and ROS 1 distros are supported.

Does the server handle custom ROS types?

Yes. It guides the LLM to discover available topics, services, actions, and their types — including custom ones — so they are used with the correct syntax without manual configuration.

What license is used?

The project is licensed under the Apache License 2.0.