Python MCP Server

What is Python MCP Server?

A Model Context Protocol (MCP) server for executing Python code and managing Python environments. It allows Claude and other LLMs to run Python code, manage files, and work with Python packages.

How to use Python MCP Server?

Ensure Node.js 16+, Python, and Claude Desktop or Cline are installed. Clone the repository, run npm install and npm run build, then add the server configuration to your Claude Desktop or Cline settings file (with the absolute path to the built dist/index.js). After restarting, you can ask the LLM to execute code, run files, list/read/write Python files, or check the Python version.

Key features of Python MCP Server

• Execute Python code snippets with optional timeout and arguments
• Execute Python files with configurable working directory
• List, read, and write Python files
• Check installed Python version and environment details
• Configurable execution timeouts, working directories, and command-line arguments

Use cases of Python MCP Server

• Run Python code directly from a chat conversation
• Execute existing Python scripts on the local machine
• Inspect or modify Python files in a project directory
• Verify Python environment details without leaving the LLM interface

FAQ from Python MCP Server

What prerequisites are needed to use Python MCP Server?

Node.js 16 or higher, Python installed and available in your PATH, and either Claude Desktop or Cline installed.

How do I fix "Python is not found" errors?

Ensure Python is installed and available in your system PATH. The server requires Python to be accessible from the command line.

What should I do if my code execution times out?

The server has a default 30‑second timeout. You can increase the timeout by specifying the timeout parameter when calling the tool.

How can I resolve permission errors when reading or writing files?

Make sure the user running the MCP server has the necessary permissions for the file paths you are accessing.

What tools are available in Python MCP Server?

The server provides six tools: execute_python_code, execute_python_file, check_python_version, list_python_files, read_python_file, and write_python_file.