https://github.com/nico0302/ros2mcp

Expose arbitrary ROS 2 services and topics as MCP tools and list topics as resources.
https://github.com/nico0302/ros2mcp

llm mcp mcp-server pixi ros ros2

Last synced: about 2 months ago
JSON representation

Expose arbitrary ROS 2 services and topics as MCP tools and list topics as resources.

• Host: GitHub
• URL: https://github.com/nico0302/ros2mcp
• Owner: Nico0302
• License: apache-2.0
• Created: 2025-07-05T12:35:10.000Z (3 months ago)
• Default Branch: main
• Last Pushed: 2025-07-27T10:12:45.000Z (2 months ago)
• Last Synced: 2025-07-27T12:11:28.283Z (2 months ago)
• Topics: llm, mcp, mcp-server, pixi, ros, ros2
• Language: Python
• Homepage:
• Size: 1.77 MB
• Stars: 1
• Watchers: 0
• Forks: 1
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

          
![ros2mcp](./assets/github_dark.svg#gh-dark-mode-only)

![ros2mcp](./assets/github_light.svg#gh-light-mode-only)

Expose arbitrary [ROS 2](https://www.ros.org/) services and topics as [MCP](https://modelcontextprotocol.io/) tools and list topics as resources.

![Example of ROS to MCP translation](./assets/ros2mcp_diagram.svg)

The `mcp_server` node can translate any ROS topic and service into a MCP resource or tool (including source comments as parameter descriptions).

[![Demo Video with TurtleBot4 and Gazebo](https://img.youtube.com/vi/yN2vLpf2-HE/0.jpg)](https://www.youtube.com/watch?v=yN2vLpf2-HE)

# Demo Setup

The demo can be run on Linux, macOS, and Windows using [Pixi](https://pixi.sh/dev/installation/).

1. Clone the repository.

```bash

git clone https://github.com/nico0302/ros2mcp.git

```

2. Start Gazebo and Rviz2 and the demo node:

```bash

pixi run demo-tb4

```

This might take a while on the first run as it downloads the necessary assets.

3. After Gazebo and Rviz2 are running, select the `2D Pose Estimate` tool in Rviz2 to set the robot's initial position by clicking in the center of the map and dragging to right:

![Rviz2 selecting 2D pose estimate](./assets/rviz.jpg)

4. Connect a MCP client to the demo server on `http://localhost:8080`

This project includes a simple MCP client that can be run with:

```bash

# make sure to set your OPENAI_API_KEY environment variable

export OPENAI_API_KEY=your_openai_api_key

pixi run demo-client

```