MCP Gs Robot

🚀 Gausium OpenAPI MCP Server

This project implements an MCP (Model Control Protocol) server, serving as a bridge to the Gausium OpenAPI. It allows AI models or other clients to interact with Gausium robots through a standardized interface.

Repository: https://github.com/cfrs2005/mcp-gs-robot

🚀 Quick Start

Environment Requirements

• Python 3.6+

Installation

pip install python-sseclient requests

Running Steps

1. Clone the repository:

   git clone https://github.com/cfrs2005/mcp-gs-robot.git
   cd mcp-gs-robot
   

2. Install dependencies:

   pip install -r requirements.txt
   

3. Set environment variables (in the terminal):

   export GS_CLIENT_ID="your_client_id"
   export GS_CLIENT_SECRET="your_client_secret"
   export GS_OPEN_ACCESS_KEY="your_access_key"
   

4. Run the server:

   python main.py
   

   By default, this command starts the server using SSE transport, binding to the address http://0.0.0.0:8000. You can modify the transport type to stdio in main.py.

✨ Features

The current server supports the following functions as MCP tools:

• list_robots: List robots accessible via the API key. (Based on: List Robots API)
• get_robot_status: Get the detailed status of a specific robot by its serial number. (Based on: Get Robot Status API)
• list_robot_task_reports: Retrieve task reports of a specific robot, filtered by time. (Based on: List Robot Task Reports API)
• list_robot_maps: Retrieve the map list of a specific robot. (Based on: List Robot Maps API)

📚 Documentation

Architecture

The server follows a layered architecture to separate concerns and facilitate maintenance:

MCP Protocol Flow

The following diagram shows how an AI model interacts with Gausium robots through the MCP protocol:

Project Structure

The layout of the code directory is as follows:

src/
├── gs_openapi/
│   ├── config.py
│   ├── client.py
│   └── tools/
│       ├── list_robots.py
│       ├── get_robot_status.py
│       ├── list_robot_task_reports.py
│       └── list_robot_maps.py
└── main.py

Project Explanation

• GS_CLIENT_ID: Your Gausium application client ID.
• GS_CLIENT_SECRET: Your Gausium application client secret.
• GS_OPEN_ACCESS_KEY: Your Gausium OpenAPI access key.

Do not hard - code credentials in the code, and do not commit them to the version control system.

Connecting MCP Clients

After the server is running, MCP clients (such as Cursor or other compatible tools) can connect to the server through an appropriate transport protocol (SSE or stdio) to use the defined tools.

Usage Example: Using with Cursor

The following diagram shows an example of how Cursor interacts with this MCP server:

Debugging

You can debug by monitoring the server logs. The basic log configuration in main.py provides timestamp, level, and source information.

Here is an example of the log output when the server is running: