MCP My Mac

MCP My Mac exposes essential Mac system information through a lightweight local API, enabling AI assistants to access real-time hardware details, configuration, and resource usage on your Mac. This enables more accurate troubleshooting and optimization while staying low-overhead and secure.

How to use

To enable your AI client to fetch Mac system details, start the MCP server locally and point your client to the provided MCP configuration. The server runs as a local process and exposes system information through an API. When your client requests data, it can query hardware specs, OS details, and resource usage to tailor responses or run targeted analyses.

How to install

Prerequisites: you need Python 3.8 or higher and the UV package manager installed on your Mac.

Step 1: Clone the MCP My Mac project repository

git clone git@github.com:zhongmingyuan/mcp-my-mac.git

Step 2: Choose how to configure the server for your MCP client

{
  "mcpServers": {
    "mcp_my_mac": {
      "command": "uv",
      "args": [
        "run",
        "-m",
        "mcp_server_my_mac"
      ]
    }
  }
}

Step 3: If you are using a Cursor-style client, add the same server by invoking the UV runner directly

uv run --directory /YOUR_PATH_TO/mcp-my-mac mcp_server_my_mac

Additional notes

The server relies on two safe, verified commands to gather information: system_profiler for hardware and software configuration details, and conda for analyzing Python environments. These commands are executed with intent to minimize risk and overhead.

Run the MCP server locally and connect your MCP client to the exposed local API. The client will automatically connect when you start it on your Mac, granting it access to the system information needed for accurate assistance with tasks like software optimization and troubleshooting.

Available tools

system_profiler

Gathers Mac system information including hardware, software, and configuration details.

conda

Analyzes Python environment configurations and dependencies.