spranab/brainstorm-mcp

Overview

An MCP server that orchestrates multi-round brainstorming debates between multiple AI models (GPT, DeepSeek, Groq, Ollama, etc.). It runs structured back-and-forth sessions where models critique and build on each other's ideas.

Best for

Best for
Developers who want to systematically compare and combine outputs from multiple AI models

Use cases

• Generate diverse solution ideas by pitting different models against each other
• Stress-test a concept through adversarial model debate
• Combine strengths of local and cloud models in a single session

How to use

Install

npm install -g brainstorm-mcp

Tools exposed

• brainstorm_quick
• brainstorm_review
• brainstorm_respond
• brainstorm_collect
• list_providers
• add_provider

Tested with

Claude Desktop, Claude Code, ChatGPT

Example client config

{\n  "mcpServers": {\n    "brainstorm": {\n      "command": "npx",\n      "args": ["-y", "brainstorm-mcp"],\n      "env": {\n        "OPENAI_API_KEY": "sk-...",\n        "GEMINI_API_KEY": "AIza...",\n        "DEEPSEEK_API_KEY": "sk-..."\n      }\n    }\n  }\n}

Notes

An MCP server that orchestrates multi-round brainstorming debates between multiple AI models (GPT, DeepSeek, Groq, Ollama, etc.). It runs structured back-and-forth sessions where models critique and build on each other’s ideas.

61 stars on GitHub. Last updated 2026-04-27. Licensed MIT.

Use cases

• Generate diverse solution ideas by pitting different models against each other
• Stress-test a concept through adversarial model debate
• Combine strengths of local and cloud models in a single session

Pros

• Leverages model diversity for richer brainstorming
• Supports both cloud and local models via Ollama
• Simple MCP integration for existing AI workflows

Cons

• Requires managing multiple API keys and endpoints
• Debate quality depends on model selection and prompt design
• Limited to text-based debate rounds, no visual or code execution

Indexed from awesome-mcp-servers-punkpeye and enriched against its public facts.