BentoRun Python MCP

Example Use Cases

• 1

  Safely execute Python code in isolated sandboxes.

• 2

  That require internet access to install packages or fetch data during code execution.

• 3

  Integration with Model Context Protocol supporting tools like Gemini CLI or Claude Code.

Description

BentoRun Python MCP provides a secure and isolated environment using gVisor sandboxes to safely execute Python code generated by AI agents supporting the Model Context Protocol. It ensures strong filesystem, resource, network, and user isolation while allowing internet access for package installation and API calls. The server supports session-based execution with preserved state and integrates with tools like Gemini CLI, Claude Code, and ADK agents.