mcp_call_conda_info
Get comprehensive information about the Conda installation on this system; with env_name, also that environment's packages and versions.
LABSCO SUMMARY
Advice about your machine, based on your machine.
Most local-model sizing advice is generic because the assistant has no idea what hardware it is talking about. This closes that gap with the smallest possible surface — two commands, read-only, nothing leaves the box. The project describes itself as beta, so treat the coverage as the starting set rather than the finished one.
What it is
A small local server that reports what your Mac actually is — hardware specifications, system configuration and current resource usage — so an assistant can answer machine-specific questions instead of guessing. It reads that state through system_profiler, and reads Python environment configuration through conda.
What you get
Your Mac's hardware specifications, system configuration and resource usage, made available to Claude Desktop or any other MCP client, plus the Python environment configuration that conda reports, and a check of whether PyTorch or TensorFlow in a given conda environment can use the GPU through Apple's Metal, with CPU-versus-GPU timings. Its source registers three tools, and all of them work through system_profiler and conda. It is aimed at people experimenting with AI and deep learning locally, where the answer to "will this model fit" depends on the actual machine.
Requirements
A Mac, Python 3.10 or later, and the UV package manager. No account and no key — everything it reads is local. Install by cloning the repository and pointing your client at it: uv --directory /YOUR_PATH_TO/mcp-my-mac run -m mcp_server_my_mac. The Python package is mcp-my-mac (0.1.0).
Setup effort
Build from source — clone the repository, then follow the steps on this page and add the configuration for your client
USE CASES
01
Deciding whether a local model will fit
Have the assistant read the actual memory, chip and storage figures before it recommends a model size or quantisation.
02
Debugging a Python environment problem
Let the assistant read the conda environment configuration directly rather than asking you to paste command output back and forth.
03
Getting optimisation advice grounded in the real machine
Ask why something is slow with the current resource usage in front of the model, instead of describing the symptoms from memory.
3 tools
Listed in project code.
mcp_call_mac_system_profiler
Call system_profiler with a given datatype (hardware, memory, displays, storage, network, power and more) to inspect the Mac's system information.
mcp_call_gpu_available
Check whether PyTorch or TensorFlow in a given conda environment can use the GPU through Apple's Metal (MPS), with CPU vs MPS benchmark timings.
BEFORE YOU INSTALL
What to know first
Tools you get
3 tools registered in the project's code.
Requires
Python 3.10 or later
Where it runs
The documented client runs the server on your machine.
INSTALLATION
Connect your client
Clone the repository, then follow the steps on this page and add the configuration for your client. This one is built from source rather than installed with a single command.
Server setup
Setup
TERMINAL
git clone https://github.com/zhongmingyuan/mcp-my-mac.gitAdd the mcpServers block below to your MCP server config file (Claude Desktop); <YOUR_PATH_TO> is the clone path
uv
TERMINAL
uv run --directory <YOUR_PATH_TO_MCP_MAC> mcp_server_my_macMCP Clients
Claude Desktop
CLAUDE DESKTOP · JSON
{
"mcpServers": {
"mcp-my-mac": {
"command": "uv",
"args": [
"--directory",
"<YOUR_PATH_TO_MCP_MAC>",
"run",
"-m",
"mcp_server_my_mac"
]
}
}
}
