About chat.mumo/mcp
chat.mumo/mcp is an MCP server published by mumo-chat in the AI category: consult a multi-model panel on contested decisions via MCP: architecture, plan review, strategy.
Recent commit activity is the strongest signal that reported issues will get attention.
Install
git clone https://github.com/mumo-chat/mumo-mcpThis server has no ConduID identity, so agent calls to it are not receipted. Pin the version you install and review the source before granting it credentials.
Tools (8)
append_roundAdd a follow-up round, optionally steered with typed snippets (KEEP / EXPLORE / CHALLENGE / CORE / SHIFT).
Parameters not published.
create_deliberationStart a deliberation. Returns an ack immediately (`session_id` + `round_id`); the models run in the background.
Parameters not published.
get_creditRead your credit balance.
Parameters not published.
get_sessionRead full session state: rounds, responses, snippets, claim maps.
Parameters not published.
list_modelsList the models available to your account, with pricing.
Parameters not published.
list_sessionsList your prior sessions.
Parameters not published.
share_sessionShare a session at its public URL; returns the page plus `.md` and `.brief.md` machine twins.
Parameters not published.
wait_for_roundBlock on cheap progress polling until the round is done, then return every model's response and the claim map.
Parameters not published.
Ask AI
Ask AI about chat.mumo/mcp
Powered by Claude · Grounded in docs
Security checks
- ✓README presentThe repository ships documentation.
5 checks have not been run yet.
README
mumo — Claude Code + Cowork plugin
Multi-model responses + cross-model reactions. Want more rounds? Context carries automatically. Stop when you have what you need.
Claude, GPT, Gemini, Grok, Qwen, GLM, Kimi in parallel. For contested decisions — architecture, plan review, pricing, strategy — where a single model might be confidently wrong.
Works in both Claude Code (developer audience) and Claude Cowork (knowledge-worker audience). Same plugin, same skill, different distribution marketplaces. For Cursor, see mumo-chat/mumo-cursor. For VS Code (GitHub Copilot), see mumo-chat/mumo-vscode.
When to use
Reach for mumo when:
- Architecture and design reviews — "Postgres or MongoDB for our event store?" "Is this migration safe under concurrent writes?"
- Plan and pre-launch pressure tests — "What would we regret 6 months in?" "Where's the biggest risk in this spec?"
- Contested product or pricing decisions — "Should this feature be paid or free?" "Which of these positionings actually converts?"
- Strategy with real cost — anywhere a confidently-wrong single model could push a team in the wrong direction for a week.
The bundled skill auto-triggers on this category and stays quiet on factual lookups, routine refactoring, and code generation from a clear spec.
What's in the box
- MCP server —
https://mumo.chat/api/mcp, seven tools:create_deliberation,wait_for_round,append_round,get_session,list_sessions,list_models,get_credit - Auto-triggering skill —
skills/mumo/SKILL.mdtells the agent when to reach for the panel so you don't have to
Install — Claude Code
Inside Claude Code:
/plugin marketplace add anthropics/claude-plugins-official
/plugin install mumo
Install — Claude Cowork
Open Claude Desktop → Cowork tab → Customize → Browse plugins → search for mumo → Install. Or browse the full catalog at claude.com/plugins.
Note: Claude Code and Cowork have separate plugin panels backed by different marketplaces. Installing in one surface does not auto-install in the other — install in each separately.
Install — VS Code (GitHub Copilot)
Install the mumo extension from the Visual Studio Marketplace. It registers the MCP server natively and stores your key in VS Code's SecretStorage — no MUMO_API_KEY env-var required.
For the full story and manual-install alternatives, see mumo-chat/mumo-vscode.
The auto-triggering skill from this repo is not bundled in the VS Code extension — Copilot's agent mode routes to the MCP server organically when you mention mumo or ask for a panel. For the explicit skill experience, use Claude Code or Cowork.
API key setup
The plugin prompts for your mumo API key at install time and stores it securely in your system keychain.
- Sign up at https://mumo.chat and create a platform key at Settings → API Keys (keys start with
mmo_live_) - Paste it when the plugin prompts you during install/enable
If you need to update the key later, reinstall or reconfigure the plugin. For manual or dev setups, you can also export MUMO_API_KEY as an environment variable.
Using the panel
Invoke the skill explicitly if the agent doesn't reach for it on its own:
- "Run this by a mumo panel"
- "Get me a second opinion from mumo on…"
- "Ask mumo about…"
Try it first
Ask Claude Code or Cowork something the panel will want to engage with:
Postgres or MongoDB for our event store given 50k events/day, a Postgres-experienced team, and a 3-month runway? What would we regret 6 months in?
The first round returns each model's prose plus a cross-model claim map showing where the panel agrees and where it splits. You can stop there, or append_round with typed snippets (KEEP / EXPLORE / CHALLENGE / CORE / SHIFT) — moderator attention on what mattered and why.
See mumo.chat/install for setup and mumo.chat/docs/mcp for the tool reference. The canonical skill lives in skills/mumo/SKILL.md — that's what Claude Code loads.
Links
- Product — https://mumo.chat
- MCP reference — https://mumo.chat/docs/mcp
- REST API — https://mumo.chat/docs/api
- Cursor plugin — https://github.com/mumo-chat/mumo-cursor
- VS Code extension — https://github.com/mumo-chat/mumo-vscode
- Issues — https://github.com/mumo-chat/mumo-mcp/issues
License
MIT
README copy captured when this server was indexed; it may lag the repository. The original is authoritative.