Rubber Duck MCP — AI skill for Claude Code

Rubber Duck MCP Server

A Model Context Protocol (MCP) server that provides a rubber duck debugging tool for LLMs. This server allows LLMs to explain their code to a "rubber duck" without expecting any response, helping them organize their thoughts and debug more effectively.

Features

• Silent Rubber Duck: A classic rubber duck debugging companion that listens without responding
• Squeaky Rubber Duck: A fun interactive rubber duck that squeaks when squeezed

Implementation

The server is implemented using FastMCP and provides two main tools:

1. rubber_duck: A traditional rubber duck debugging tool that listens silently
2. squeak: An interactive rubber duck that responds with a "Squeak!" when activated

Usage

The server can be used by connecting to it through the MCP protocol. It's particularly useful for:

• Debugging complex code issues
• Walking through implementation logic
• Organizing thoughts during development
• Adding a bit of fun to your debugging process with the squeaky duck feature

Installation

Requirements

• Python 3.10 or higher
• uv package manager
• fastmcp package

To use this rubber duck server with Claude, you'll need to install it using the FastMCP CLI. Run the following command:

fastmcp install src/server.py

This will make rubber-duck available to Claude through its MCP configuration file `claude_desktop_config.json`. For more details about Claude Desktop integration, see the [FastMCP documentation](https://github.com/jlowin/fastmcp/blob/main/README.md#claude-desktop-integration-for-regular-use).

The install command will generate a json object of this format:

{
    "mcpServers": {
      "rubber-duck": {
        "command": "uv",
        "args": [
          "run",
          "--with",
          "fastmcp",
          "fastmcp",
          "run",
          "/src/server.py"
        ]
      }
    }
}

To use rubber-duck in Cursor, add this json to `/.cursor/mcp.