Brainstorm MCP
Multi-provider brainstorming via MCP — GPT-5, DeepSeek, Claude side-by-side.
https://github.com/spranab/brainstorm-mcpverification
L1Ownership verified · GitHub repo admin: spranab/brainstorm-mcp
Ownership verified — not quality endorsement. icantmarket verifies control of listed assets only. last rechecked 9/3/2026.
stars
71
forks
8
created
Feb 2026
pushed
Sep 2026
active ask · feedback
The synthesis output is a 3-bullet Recommendation / Key Tradeoffs / Strongest Disagreement verdict. Is that scannable enough when reading cold, or does the actual debate transcript drown the conclusion below it?
- Looking for:
- Anyone who has used multi-model tools (Cursor agent mode, Codex CLI, GPT-5 agentic) and reads model output regularly.
- Time:
- 15m
- In return:
- I will run brainstorm-mcp on a design question you have — pick the topic, I will send you the synthesis.
- Closes:
- 6/16/2026
Success means: I learn whether the synthesis-first-then-transcript format reads better than transcript-first-then-synthesis.
Posts · 1
Recent activity · 5
GitHub releases · PyPI downloads · npm versions
- release
**Support for the latest models; fixes defaults that had stopped working.** - **Default models updated:** `openai` → `gpt-6.1-sol`, `gemini` → `gemini-3.8-flash`, `deepseek` → `deepseek-flash`, `m
github releases · 9/30/2026, 12:15:01 AM
- release
**GPT-6 support.** OpenAI's GPT-6 ships under the model id `gpt-6-astra` (there is no plain `gpt-6`). It takes the same request shape as gpt-5.x: `max_completion_tokens`, default temperature only. br
github releases · 9/7/2026, 5:15:02 PM
- release
v1.6.0 — CLI providers: debate on a subscription instead of API credits
**CLI providers — debate on a subscription instead of API credits.** A third transport alongside API and hosted mode. brainstorm can now shell out to agent CLIs installed on your machine, so debates
github releases · 8/14/2026, 6:15:01 AM
- release
v1.5.4 — MCP Registry submission
Add mcpName field for Official MCP Registry publishing.
github releases · 6/2/2026, 7:15:03 AM
- release
v1.5.6 — gpt-5.x temperature fix
## Fix Drop `temperature` field for gpt-5.x reasoning models. The `useNewTokenParam` branch already correctly used `max_completion_tokens` for these models, but it still passed `temperature: 0.7` wh
github releases · 6/2/2026, 7:15:03 AM
Audit log · 5
Public · append-only
- 9/3/2026, 4:00:02 AMreverifiedtrust=L1
- 8/3/2026, 4:00:02 AMreverifiedtrust=L1
- 7/3/2026, 4:00:03 AMreverifiedtrust=L1
- 6/2/2026, 6:42:20 AMverifiedtrust=L1
- 6/2/2026, 6:42:20 AMclaimedtrust=L1