MATLAB MCP Core Server

Controls MATLAB from AI apps by starting MATLAB, executing code, and validating MATLAB scripts.

other

80

GitHub Stars

other

Language

6 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "matlab-matlab-mcp-core-server": {
      "command": "claude",
      "args": [
        "mcp",
        "add",
        "--transport",
        "stdio",
        "matlab",
        "/fullpath/to/matlab-mcp-core-server-binary",
        "[arguments...]"
      ]
    }
  }
}

You run the MATLAB MCP Core Server to start and control MATLAB from AI applications, enabling you to start and quit MATLAB, execute MATLAB code, and assess code quality from your AI workflows.

How to use

You integrate the MATLAB MCP Core Server with your MCP client to access MATLAB capabilities from your AI tools. You can add the server as a local stdio server or connect through Claude Code, Claude Desktop, or GitHub Copilot in Visual Studio Code by pointing your MCP client at the MATLAB MCP Core Server binary and providing any optional arguments. The server runs as a local process and communicates via stdio, accepting MATLAB commands, scripts, and code evaluations as part of your workflow.

How to install

Prerequisites: ensure you have MATLAB (MathWorks) 2020b or later installed and that MATLAB is discoverable on your system PATH. You will also need an environment capable of running the MCP client you choose (for example Claude Code, Claude Desktop, or VS Code with the MCP extension). The following installation steps cover Windows, Linux, and macOS.

# Windows / Linux / macOS (MATLAB MCP Core Server)

# 1) Install MATLAB (2020b or later) and ensure it is on PATH
# 2) For Windows or Linux, download the Latest Release and run the server binary, or build from source if you prefer:
#    go install github.com/matlab/matlab-mcp-core-server/cmd/matlab-mcp-core-server@latest

# macOS (Apple silicon)
#  curl -L -o ~/Downloads/matlab-mcp-core-server https://github.com/matlab/matlab-mcp-core-server/releases/latest/download/matlab-mcp-core-server-maca64
# macOS (Intel)
#  curl -L -o ~/Downloads/matlab-mcp-core-server https://github.com/matlab/matlab-mcp-core-server/releases/latest/download/matlab-mcp-core-server-maci64
# Then grant executable permissions
#  chmod +x ~/Downloads/matlab-mcp-core-server

# 3) Add the server to your MCP client configuration as described by your client (Claude Code, Claude Desktop, VS Code, etc.)

Configuration examples for common clients

Below are examples showing how to connect the MATLAB MCP Core Server to different MCP clients. Adapt paths and options to your environment.

# Claude Code: add a local stdio server for MATLAB
claude mcp add --transport stdio matlab /fullpath/to/matlab-mcp-core-server-binary [arguments...]

# Example with an initial working folder
claude mcp add --transport stdio matlab /fullpath/to/matlab-mcp-core-server-binary --initial-working-folder=/home/username/myproject

GitHub Copilot in Visual Studio Code

In VS Code, add a local stdio server for MATLAB using the MCP configuration in your workspace. Provide the full path to the server binary and any arguments.

{
  "servers": {
    "matlab": {
      "type": "stdio",
      "command": "/fullpath/to/matlab-mcp-core-server-binary",
      "args": []
    }
  }
}

Available tools

detect_matlab_toolboxes

Lists installed MATLAB toolboxes with version information.

check_matlab_code

Performs static code analysis on a MATLAB script, returning warnings about style, potential errors, and best practices without executing the script.

evaluate_matlab_code

Evaluates a string of MATLAB code and returns the output.

run_matlab_file

Executes a MATLAB script and returns the output.

run_matlab_test_file

Executes a MATLAB unit test script and returns comprehensive test results.