mcp-python-interpreter

概述

MCP Python 解释器是一个功能强大的服务器，使 AI 模型能够与 Python 环境交互。它提供了在不同 Python 环境中执行代码、管理软件包和执行文件操作的综合功能。该服务器通过隔离的工作目录和文件路径限制实施了安全措施，使其适用于开发工作流程。文档详尽，包含清晰的示例和安全考虑说明。

试试问 AI

装完之后，这里有 5 个你可以让 AI 做的事：

• list_python_environments

  List all available Python environments (system and conda)

  试试:Show me all available Python environments on my system试试:List all conda environments
• list_installed_packages

  List packages installed in a specific Python environment

  试试:List all packages installed in my system Python environment试试:Show me the packages in my conda-base environment
• install_package

  Install a Python package in a specific environment

  试试:Install the requests package in my system Python environment试试:Add pandas to my conda-base environment
• run_python_code

  Execute Python code in a specific environment

  试试:Run this Python code in my conda-base environment: print('Hello, world!')试试:Calculate 2+2 in my system Python
• run_python_file

  Execute a Python file in a specific environment

  试试:Run data_analysis.py with these arguments: --input=data.csv --output=results.csv试试:Execute my script.py file
• read_file

  Read contents of any file type, with size and safety limits

  试试:Read the contents of my 'data.json' file试试:Show me what's in my analysis.py file
• write_file

  Create or overwrite files with text or binary content

  试试:Write a new configuration file with these settings...试试:Create a text file with 'Hello World' content
• write_python_file

  Create or overwrite a Python file specifically

  试试:Create a new Python file called 'hello.py' with a function that says hello试试:Write a Python script to process my data
• list_directory

  List Python files in a directory

  试试:Show me all Python files in my current directory试试:List the .py files in my project folder
• python_function_template

  Generate a template for a Python function

  试试:Create a template for a Python function that processes data试试:Generate a function template for API calls
• refactor_python_code

  Help refactor Python code

  试试:Refactor this Python function to be more efficient试试:Improve the readability of this code
• debug_python_error

  Help debug a Python error

  试试:Debug this Python error: NameError: name 'x' is not defined试试:Fix the syntax error in my code

可对比工具

安装

pip install mcp-python-interpreter

uv install mcp-python-interpreter

{
  "mcpServers": {
    "mcp-python-interpreter": {
        "command": "uvx",
        "args": [
            "mcp-python-interpreter",
            "--dir",
            "/path/to/your/work/dir",
            "--python-path",
            "/path/to/your/python"
        ],
        "env": {
            "MCP_ALLOW_SYSTEM_ACCESS": 0
        },
    }
  }
}

FAQ

采取了哪些安全措施？

服务器在通过 --dir 参数指定的隔离工作目录中运行，具有文件大小限制（最多 1MB），防止在指定目录外写入，并包含显式的覆写保护。

我可以使用不同的 Python 环境吗？

是的，服务器列出了允许使用的不同 Python 环境，包括系统环境和 conda 环境，可通过 list_python_environments 工具访问。

mcp-python-interpreter 对比