MATLAB

🚀 MATLAB MCP Server

This Model Context Protocol (MCP) server provides integration with MATLAB, allowing you to create and execute MATLAB scripts and functions via Claude or other MCP clients.

🚀 Quick Start

This MCP server enables seamless integration with MATLAB, facilitating the creation and execution of MATLAB scripts and functions through Claude or other MCP clients.

✨ Features

The server offers the following tools:

1. create_matlab_script: Create a new MATLAB script file.

   • Scripts are saved in the matlab_scripts directory.
   • The file name must be a valid MATLAB identifier.

2. create_matlab_function: Create a new MATLAB function file.

   • Functions are saved in the matlab_scripts directory.
   • The file name must follow MATLAB function naming rules.

3. execute_matlab_script: Execute a MATLAB script.

   • Run the script by specifying the script path and parameters.

4. call_matlab_function: Call a MATLAB function.

   • Call the function by specifying the function name and input parameters.

📦 Installation

Installation Requirements

• Python 3.11 (Python 3.13 and 3.12 are not supported at present)
• MATLAB R2024a (or a compatible version)
• uv package manager

Installation Steps

1. Create and configure a Python environment:

   # Lock the Python version
   uv python pin 3.11
   
   # Create a virtual environment
   uv venv
   
   # Activate the virtual environment
   source .venv/bin/activate
   
   # Install MCP
   uv add "mcp[cli]"
   

2. Install the MATLAB engine: The MATLAB engine will be automatically installed when the server runs for the first time, using the MATLAB installation specified by the MATLAB_PATH environment variable.

💻 Usage Examples

Directory Structure

• matlab_server.py: The main implementation file of the MCP server
• matlab_scripts/: All MATLAB scripts and functions are saved in this directory (automatically created)
• pyproject.toml: Python project configuration file
• .python-version: File specifying the Python version

Claude Desktop Integration

1. Open your Claude desktop configuration:

   # On macOS
   code ~/Library/Application\ Support/Claude/claude_desktop_config.json
   

2. Add the MATLAB server configuration:

   {
       "mcpServers": {
           "matlab": {
               "command": "uv",
               "args": [
                   "--directory",
                   "/absolute/path/to/matlab-mcp",
                   "run",
                   "matlab_server.py"
               ],
               "env": {
                   "MATLAB_PATH": "/Applications/MATLAB_R2024a.app"
               }
           }
       }
   }
   

   ⚠️ Important Note

   • Replace /absolute/path/to/matlab-mcp with the absolute path of your actual project directory.
   • Verify that MATLAB_PATH points to your MATLAB installation location.
   • Use absolute paths (instead of relative paths).

Testing and Verification

1. Create a MATLAB script file (e.g., test_script.m):

   function result = test_function(input)
       result = input + 1;
   end
   

2. Run the test script on the server:

   uv run matlab_server.py --test
   

📚 Documentation

Script Storage

• All MATLAB scripts and functions are saved in the matlab_scripts directory.
• Each script or function file should have a unique name and follow MATLAB's naming conventions.

Environment Variables

• MATLAB_PATH: Specifies the location of the MATLAB installation directory, with a default value of /Applications/MATLAB_R2024a.app.

Troubleshooting

1. Failed MATLAB engine installation

   • Check if MATLAB_PATH is correct.
   • Ensure that MATLAB is installed correctly and has executable permissions.

2. Incompatible Python version

   • Verify that the Python version is 3.11.
   • Use the uv python version command to check the current Python version.

3. Scripts or functions cannot be executed

   • Check if there are syntax errors in the script or function file.
   • Ensure that the script or function has executable permissions.

Update and Maintenance

1. Regularly check for the latest versions of MATLAB and Python.
2. Use the uv update command to update the MCP server and related dependencies.
3. When encountering issues, refer to the official documentation or contact technical support.