MCP Catalogs
Homeipybox screenshot

ipybox

by gradion-ai·74·Score 46

Unified execution environment for Python, shell commands, and MCP tool calls with IPython kernels.

developer-toolsai-llmproductivity
19
Forks
4
Open issues
2 mo ago
Last commit
2d ago
Indexed

Overview

ipybox is a comprehensive execution environment that allows stateful code execution through IPython kernels, enabling seamless integration of Python code, shell commands, and programmatic MCP tool calls in a single environment. It features optional lightweight sandboxing through Anthropic's sandbox-runtime, ensuring security while maintaining local access to data and tools. The project generates Python APIs for MCP server tools using mcpygen and provides application-level approval mechanisms for tool calls and shell commands during execution.

Try asking AI

After installing, here are 5 things you can ask your AI assistant:

you:Interactive development environment combining Python, shell commands, and MCP tools
you:Sandboxed code execution for secure data processing with local access
you:Building Freeact agents with unified programmatic tool access
you:What is ipybox?
you:How does ipybox differ from traditional IPython?

When to choose this

Choose ipybox when you need a unified environment for Python, shell commands, and MCP tools with sandboxing and approval controls, especially for building secure AI applications.

When NOT to choose this

Avoid ipybox if you need pure JavaScript/TypeScript execution environments or if you're looking for a managed cloud service rather than a local execution solution.

Tools this server exposes

4 tools extracted from the README
  • execute_python

    Execute Python code in a stateful IPython kernel

  • execute_shell

    Execute shell commands with captured output

  • mcp_tool_call

    Call MCP tools programmatically via generated Python API

  • create_code_action

    Create code actions for modifying files or content

Note: Tools inferred from the capabilities section which mentions programmatic MCP tool calls and shell command execution, but exact tool names are not explicitly listed in the README.

Comparable tools

jupyter-mcppython-mcpsandbox-mcp

Installation

Installation

pip install ipybox

Claude Desktop Configuration

Add to your Claude Desktop configuration:

{
  "mcpServers": {
    "ipybox": {
      "command": "python",
      "args": ["-m", "ipybox.mcp_server"]
    }
  }
}

FAQ

What is ipybox?
ipybox is a unified execution environment that combines Python code, shell commands, and programmatic MCP tool calls in stateful IPython kernels with optional sandboxing.
How does ipybox differ from traditional IPython?
ipybox extends traditional IPython by adding programmatic MCP tool calls, approval mechanisms for commands, and optional sandboxing for enhanced security.

Compare ipybox with

GitHub →

Last updated · Auto-generated from public README + GitHub signals.