Safe Local Python Executor MCP server

Safe Local Python Executor is an MCP server that wraps Hugging Face's LocalPythonExecutor to provide a secure environment for running Python code generated by LLMs. It acts as a code interpreter tool for LLM applications like Claude Desktop, enabling you to safely execute Python code without requiring Docker or a VM.

Features

• Provides a run_python tool accessible via MCP (Model Context Protocol)
• Offers safer execution compared to direct use of Python eval()
• Runs in a Python virtual environment using uv
• Restricts file I/O operations
• Limits imports to a safe subset:
  • collections
  • datetime
  • itertools
  • math
  • queue
  • random
  • re
  • stat
  • statistics
  • time
  • unicodedata

Installation Options

Via Smithery

The simplest way to install Safe Local Python Executor for Claude Desktop is via Smithery:

npx -y @smithery/cli install @maxim-saplin/mcp_safe_local_python_executor --client claude

Manual Installation

1. Install uv package manager:

   # macOS
   brew install uv
   
   # For other platforms, follow the official installation guide:
   # https://docs.astral.sh/uv/getting-started/installation/
   

2. Clone the repository and navigate to it:

   git clone https://github.com/your-username/mcp_safe_local_python_executor
   cd mcp_safe_local_python_executor
   

3. Start the server:

   uv run mcp_server.py
   

   A virtual environment will be created automatically and all dependencies (smolagents, mcp) will be installed.

Configuring Claude Desktop

1. Install Claude Desktop from claude.ai

2. Edit your Claude Desktop configuration file:

   • macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
   • Windows: %APPDATA%\Claude\claude_desktop_config.json
   • Or navigate to Claude Desktop → Settings → Developer → click "Edit Config" button

3. Add the following configuration:

{
    "mcpServers": {
        "safe-local-python-executor": {
            "command": "uv",
            "args": [
                "--directory", 
                "/path/to/mcp_local_python_executor/",
                "run",
                "mcp_server.py"
            ]
        }
    }
}

4. Replace /path/to/mcp_local_python_executor/ with the actual path to your installation

5. Restart Claude Desktop

6. The Python executor tool will now be available in Claude (look for the hammer icon in the message input field)

Usage Examples

Once configured, you can use prompts like:

• "Calculate the factorial of 5 using Python"
• "Create a list of prime numbers up to 100"
• "Solve this equation using Python: x^2 + 5x + 6 = 0"

When you use these prompts, Claude will use the safe Python executor to run the code and return the results.

Security Considerations

This executor provides a balance between convenience and security. While not as secure as running code in a Docker container or VM, it's significantly safer than using direct eval() calls by:

• Restricting imports to a safe subset of modules
• Preventing file I/O operations
• Running in an isolated Python environment

For more information on security aspects, visit the Hugging Face documentation.

How to add this MCP server to Cursor

There are two ways to add an MCP server to Cursor. The most common way is to add the server globally in the ~/.cursor/mcp.json file so that it is available in all of your projects.

If you only need the server in a single project, you can add it to the project instead by creating or adding it to the .cursor/mcp.json file.

Adding an MCP server to Cursor globally

To add a global MCP server go to Cursor Settings > MCP and click "Add new global MCP server".

When you click that button the ~/.cursor/mcp.json file will be opened and you can add your server like this:

{
    "mcpServers": {
        "cursor-rules-mcp": {
            "command": "npx",
            "args": [
                "-y",
                "cursor-rules-mcp"
            ]
        }
    }
}

Adding an MCP server to a project

To add an MCP server to a project you can create a new .cursor/mcp.json file or add it to the existing one. This will look exactly the same as the global MCP server example above.

How to use the MCP server

Once the server is installed, you might need to head back to Settings > MCP and click the refresh button.

The Cursor agent will then be able to see the available tools the added MCP server has available and will call them when it needs to.

You can also explictly ask the agent to use the tool by mentioning the tool name and describing what the function does.