mcp-rubber-duck

general MCP Server

An MCP server that acts as a bridge to query multiple OpenAI-compatible LLMs with MCP tool access. Just like rubber duck debugging, explain your problems to various AI "ducks" who can actually research and get different perspectives!

Install Ready
generalgeneral
3 views168 stars24 forksv1.19.2MIT

Why This Matters

Discovered via unknown and last synced 3mo ago.

Install Ready
Source
unknown
Stars
168
Last synced
3mo ago
Install
Instructions detected

Install

1. Install the package

npx mcp-rubber-duck

2. Add to claude_desktop_config.json

{
  "mcpServers": {
    "mcp-rubber-duck": {
      "command": "npx",
      "args": [
        "mcp-rubber-duck"
      ]
    }
  }
}

Config file location: ~/Library/Application Support/Claude/claude_desktop_config.json (macOS) / %APPDATA%\Claude\claude_desktop_config.json (Windows)

38
Tools
0
Resources
0
Prompts
Standard I/O
Transport

Available Tools (38)

DEFAULT_TEMPERATURE

Default temperature (e.g., `0.7`)

list_ducks

List configured providers and health status

duck_debate

Structured multi-round debate between ducks

MCP_SERVER

Set to `true` for MCP server mode

compare_ducks

Ask the same question to multiple providers simultaneously

get_pending_approvals

Pending MCP tool approval requests

Variable

Description

OPENAI_API_KEY

OpenAI API key

GEMINI_API_KEY

Google Gemini API key

GROQ_API_KEY

Groq API key

DEFAULT_PROVIDER

Default provider (e.g., `openai`)

LOG_LEVEL

`debug`, `info`, `warn`, `error`

MCP_BRIDGE_ENABLED

Enable MCP Bridge (ducks access external MCP servers)

duck_council

Get responses from all configured ducks

approve_mcp_request

Approve or deny a duck's MCP tool request

Tool

Description

get_usage_stats

Usage statistics and estimated costs

Prompt

Purpose

diverge_converge

Divergent exploration then convergence

ask_duck

Ask a single question to a specific LLM provider

duck_vote

Multi-duck voting with reasoning and confidence

perspectives

Multi-angle analysis with assigned lenses

chat_with_duck

Conversation with context maintained across messages

duck_judge

Have one duck evaluate and rank others' responses

assumptions

Surface hidden assumptions in plans

Guardrails

[docs/guardrails.md](./docs/guardrails.md)

clear_conversations

Clear all conversation history

list_models

List available models for providers

duck_iterate

Iteratively refine a response between two ducks

blindspots

Hunt for overlooked risks and gaps

mcp_status

MCP Bridge status and connected servers

red_team

Security/risk analysis from multiple angles

tradeoffs

Structured option comparison

Roadmap

[docs/roadmap.md](./docs/roadmap.md)

reframe

Problem reframing at different levels

architecture

Design review across concerns

Topic

Link

Architecture

[docs/architecture.md](./docs/architecture.md)