opp-repl-mcp-server

Execute Python code in a live opp_repl IPython session via MCP.

Updated Apr 23, 2026
One-click install
npx skills add https://github.com/tabgab/opp_repl-skill --skill opp-repl-mcp-server
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: opp-repl-mcp-server
Source: https://github.com/tabgab/opp_repl-skill/tree/main/opp-repl-mcp-server
Command: npx skills add https://github.com/tabgab/opp_repl-skill --skill opp-repl-mcp-server

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This MCP server enables AI agents to execute Python code inside a live opp_repl IPython session and explore opp_repl resources over a streamable HTTP MCP interface.

Core Features & Use Cases

  • Execute arbitrary Python code in the live IPython kernel via execute_python.
  • Browse opp_repl resources (guides, packages, classes, methods, functions) through opp_repl:// URIs.
  • Load and expose an MCP endpoint that integrates with Claude Desktop, Claude Code, Cursor, or custom agents.

Quick Start

Start the MCP server alongside the REPL using the default port.

Frequently Asked Questions about opp-repl-mcp-server

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I execute Python code in a live IPython kernel from an AI agent?

You can execute Python code in a live IPython kernel from an AI agent by connecting to an MCP server that exposes an execute_python tool over a streamable HTTP endpoint for arbitrary code execution.

Can I use an MCP server to integrate opp_repl with Claude Desktop or Cursor?

Yes, you can integrate opp_repl with Claude Desktop, Claude Code, or Cursor by loading the MCP server endpoint, allowing the AI agent to access live execution and documentation resources directly.

How do I browse opp_repl guides and API documentation through an MCP interface?

You can browse opp_repl guides, packages, classes, methods, and functions through opp_repl:// URIs exposed by the MCP server, providing direct access to API documentation resources over HTTP.

What port does the opp_repl MCP server use by default?

The opp_repl MCP server defaults to port 9966 for its streamable HTTP endpoint at /mcp, but this port is configurable to support different network environments.

Do I need any external dependencies to run the opp_repl MCP server?

No external dependencies are required to run the opp_repl MCP server, as it operates independently to expose the live IPython session and documentation resources to connected AI agents.