Python MCP

🚀 Local Python MCP Server

A MCP server that provides an interactive Python command-line environment (Read-Eval-Print Loop).

🚀 Quick Start

This is a MCP server that offers an interactive Python command-line environment (REPL). It provides access to REPL session history and implements a tool to execute Python code in persistent sessions.

✨ Features

Resources

The server provides access to the REPL session history:

• A custom repl:// URI scheme is used to access the session history.
• The history of each session can be viewed as a plain text resource.
• The history shows each executed input code and its corresponding output.

Tools

The server implements a tool:

• python_repl: Executes Python code in a persistent session.
  • Accepts code (Python code to execute) and session_id as required parameters.
  • Each session maintains an independent state.
  • Supports expressions and statements.
  • Captures and returns stdout/stderr output.

📦 Installation

Configuration

Claude Desktop

On MacOS: ~/Library/Application\ Support/Claude/claude_desktop_config.json On Windows: %APPDATA%/Claude/claude_desktop_config.json%

📚 Documentation

Build and Publish

To prepare the distribution package:

1. Synchronize dependencies and update the lock file:

uv sync

2. Build the package distribution:

uv build

This will create source and wheel format distributions in the dist/ directory.

3. Publish to PyPI:

uv publish

Note: You need to set your PyPI credentials via environment variables or command-line parameters:

• Token: --token or UV_PUBLISH_TOKEN
• Or username/password: --username/UV_PUBLISH_USERNAME and --password/UV_PUBLISH_PASSWORD

Debugging

Since the MCP server runs via standard input and output, debugging can be challenging. For the best debugging experience, we highly recommend using MCP Inspector.

You can start the MCP Inspector using this command via npm:

npx @modelcontextprotocol/inspector uv --directory /path/to/python_local run python-local

After starting, the Inspector will display a URL that you can access in your browser to start debugging.