MCP My Mac

MCP My Mac is an MCP server that exposes macOS system hardware specifications, system configurations, and environment details to AI assistants. Built for developers and researchers experimenting with AI and deep learning workflows on Mac machines, the server provides an interface to query local machine states directly from MCP-enabled clients like Claude Desktop and Cursor. The server operates locally with minimal overhead and executes verified system commands rather than arbitrary scripts, restricting its operations to safe utilities like system_profiler for hardware diagnostics and conda for evaluating Python environment setups. By connecting Claude or another LLM client to local system profiles, users eliminate the need to manually copy-paste terminal outputs, memory limits, chip architectures, or package configurations during debugging sessions. This enables language models to generate machine-specific optimization advice, verify dependency setups, troubleshoot GPU or memory constraints for local model training, and assist with environment configurations based on accurate, real-time operating system context.

Category: Cloud & Infrastructure

Tags: hardware, macos, monitoring, system-info

Visit MCP My Mac

How to install and configure MCP My Mac

1. Ensure Python 3.8 or higher and the uv package manager are installed on your Mac. 2. Clone the repository to your local machine: bash git clone git@github.com:zhongmingyuan/mcp-my-mac.git 3. To configure for Claude Desktop, add the server to your claude_desktop_config.json: json { "mcpServers": { "mcp-my-mac": { "command": "uv", "args": [ "--directory", "/YOUR_PATH_TO/mcp-my-mac", "run", "-m", "mcp_server_my_mac" ] } } } Replace /YOUR_PATH_TO with your actual cloned path. 4. For Cursor, add a new tool using the command type: bash uv run --directory /YOUR_PATH_TO/mcp-my-mac mcp_server_my_mac 5. Restart your client to establish the connection.

What you can do with MCP My Mac

• Providing your hardware specs, Apple silicon details, and memory limits directly to AI assistants for local deep learning troubleshooting. - Analyzing Conda environment setups and installed dependencies without manually copying package lists into the chat window. - Diagnosing system resource constraints and bottlenecks when running heavy compilation tasks or local machine learning models. - Assisting AI coding assistants in tuning compiler flags, framework installations, and execution scripts to match your exact macOS version.

Key facts

• Open Source
• https://github.com/zhongmingyuan/mcp-my-mac
• Cloud & Infrastructure, Developer Tools & Code Intelligence
• hardware, macos, monitoring, system-info

Part of MCP Servers

What is MCP My Mac?

MCP My Mac is a Model Context Protocol server that allows AI assistants like Claude Desktop to access local macOS hardware details, system configurations, and Conda environment information through a controlled local interface.

What commands does MCP My Mac run on my machine?

The server runs only verified, safe system inspection commands locally. Specifically, it executes system_profiler to gather macOS system and hardware specifications and conda to inspect Python environment configurations.

Which MCP clients work with MCP My Mac?

MCP My Mac works with any client supporting the Model Context Protocol over stdio. The official documentation specifically details configuration steps for Claude Desktop and Cursor.

What prerequisites are required before installing MCP My Mac?

You need a Mac running Python 3.8 or higher, Git to clone the repository, and the uv package manager installed to run the local server process.

Is MCP My Mac open source?

Yes, MCP My Mac is an open-source project hosted on GitHub under the repository zhongmingyuan/mcp-my-mac.