spranab/brainstorm-mcp

Ask one model a design question and you get one confident answer, with no
signal about which parts it is unsure of. Ask three and the disagreement is
the signal.
brainstorm-mcp runs multi-round debates between GPT, Gemini, DeepSeek, Claude
and local Ollama models from inside your editor: they see and critique each
other's answers across rounds, then you get a 3-bullet synthesis —
recommendation, key tradeoffs, strongest disagreement. Also does instant
quick mode, multi-model code review with verdicts, and red-team/Socratic
styles. Hosted mode needs zero API keys.
**Don't trust one AI. Make them argue.**
Quick start (60 seconds)
Then ask Claude: *"Brainstorm using opus, sonnet, and haiku about whether we should use GraphQL or REST."*
That runs in hosted mode: no API keys, no provider accounts. The debate uses the models already
available in your environment, and you get the 3-bullet synthesis at the end.
Demo
> Click to watch: 3 models debate, cross-examine, and produce a structured verdict — all inside Claude Code.
Features
- **Hosted mode** — No API keys needed. Uses models in your environment (Claude Opus/Sonnet/Haiku) via sub-agents
- **API mode** — Direct model API calls with parallel execution across OpenAI, Gemini, DeepSeek, Groq, Ollama
- **CLI mode** — Debate through agent CLIs you already have (claude, codex, and more) so debates run on your **subscription instead of API credits**
- **brainstorm_quick** — Instant multi-model perspectives in under 10 seconds
- **brainstorm_review** — Multi-model code review with structured findings, severity ratings, and verdicts
- **Debate styles** — Freeform, red-team (adversarial), and Socratic (probing questions)
- **Context injection** — Ground debates in actual code, diffs, or architecture docs
- **3-bullet synthesis verdicts** — Recommendation, Key Tradeoffs, Strongest Disagreement
- **Claude as participant** — Claude debates alongside external models with full conversation context
- **Multi-round debates** — Models see and critique each other's responses across rounds
- **Parallel execution** — All models respond concurrently within each round
- **Resilient** — One model failing doesn't abort the debate
- **Cross-platform** — Works on macOS, Windows, and Linux
Install
The one-line command under Quick start is enough for hosted mode. Add provider keys to
bring GPT, Gemini, DeepSeek, Groq or Ollama into the debate; per-client setup follows.
Claude Code
Add to your project's .mcp.json:
Claude Desktop
Add to claude_desktop_config.json:
Manual Install
> **Hosted mode** requires no API keys — just install and go. The host (Claude Code) executes prompts using its own model access.
Configuration
Option 1: Environment Variables (simplest)
Option 2: Config File (full control)
Set BRAINSTORM_CONFIG to point to a JSON config:
Known providers (openai, gemini, deepseek, groq, mistral, together, moonshot,
minimax, glm, qwen, xai) don't need a baseURL, and the ones with a built-in default model
(openai, gemini, deepseek, moonshot, minimax, glm, qwen, xai) don't need a model either.
Use provider:default (e.g. openai:default) to mean a provider's default model.
Env-var detection: OPENAI_API_KEY, GEMINI_API_KEY, DEEPSEEK_API_KEY, MOONSHOT_API_KEY,
MINIMAX_API_KEY, ZAI_API_KEY, DASHSCOPE_API_KEY (Qwen), XAI_API_KEY (Grok; GROK_API_KEY also works). Each also accepts
_BASE_URL and _DEFAULT_MODEL.
| Provider | Default model |
|----------|---------------|
| openai | gpt-6.1-sol (also gpt-6-astra, gpt-6-sol, gpt-6-luna, gpt-5.5) |
| gemini | gemini-3.8-flash (gemini-3.1-pro-preview for Pro) |
| deepseek | deepseek-flash (deepseek-v4-pro) |
| xai | grok-4.7 |
| moonshot | kimi-k3 |
| minimax | MiniMax-M3 |
| glm | glm-5.3 |
| qwen | qwen3.8-max |
Reasoning models (GPT-5/6, o-series, Kimi K3, Gemini 3, DeepSeek V4, Grok 4+, GLM-5, MiniMax M3,