# Rubber Duck

> Use this tool when you need to leverage multiple large language models (LLMs) for enhanced decision-making and conversation management. It solves problems of model bias and reliability by querying multiple LLMs simultaneously and using consensus voting to determine the most accurate response. It takes in natural language inputs and outputs ranked responses with confidence scores, making it ideal for applications requiring robust and reliable language understanding.

Canonical page: https://skillsregistry.net/skills/nesquikm-rubber-duck  
JSON: https://api.skillsregistry.net/v1/skills/nesquikm-rubber-duck

## Description

Bridges to multiple OpenAI-compatible LLM providers including OpenAI, Google Gemini, Groq, Together AI, Perplexity, and Ollama. Features a duck council that queries all configured LLMs simultaneously, consensus voting with reasoning and confidence scores, and LLM-as-Judge evaluation where models rank each other's responses. Supports conversation management, response caching, and automatic failover.

## Trust

- **Trust score (0–1):** 0.69
- **Verification tier:** scanned
- **Last scanned:** 2026-09-19

## Facts

- **Version:** 1.0.0
- **Skill type:** atomic
- **Execution layer:** mcp-remote
- **Runtime environment:** api
- **Category:** ai-ml
- **Updated:** 2026-09-19

## Source

- **Source listing:** [PulseMCP](https://www.pulsemcp.com/servers/nesquikm-rubber-duck)
- **Repository:** <https://github.com/nesquikm/mcp-rubber-duck>

## Use it

Resolve this record through the SkillsRegistry MCP server (no auth, read-only):

```
claude mcp add --transport http --scope user skillsregistry https://api.skillsregistry.net/mcp
```

```json
{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "get_skill",
    "arguments": {
      "slug": "nesquikm-rubber-duck"
    }
  }
}
```

REST: `GET https://api.skillsregistry.net/v1/skills/nesquikm-rubber-duck` · pull for local use: `GET https://api.skillsregistry.net/v1/skills/nesquikm-rubber-duck/pull`

---
SkillsRegistry indexes agent skills from public registries and GitHub. Skills we have analysed are scanned with Circle-IR and scored on six dimensions; each listing states its scan coverage. More: https://skillsregistry.net/llms.txt