How mumo works

Your agent already does the work.
mumo is for the calls it shouldn't make alone.

When you hit something complex, contested, or expensive to get wrong, your agent runs a panel of 2–3 models over your private context — and gets back where they agree, where they break, and why. Your agent does the synthesis; you get the audit trail when it matters. The disagreement is the deliverable, not a step on the way to one answer.

When you reach for it

Use a synthesizer for facts. Use mumo for decisions.

A compound model that collapses a panel into one answer is genuinely good — for the work where there's a right answer to find. mumo is for the other kind: the call that's contested, expensive to get wrong, and specific to your situation.

Routine / verifiable
One good model

Code, lookups, well-trodden tasks with an answer that isn't in dispute. A single frontier model is faster and cheaper.

→ your daily driver
Research / breadth
Synthesis

Wide public retrieval where coverage is the bottleneck and a best answer exists to converge on. The intern's executive summary.

→ a compound model
Contested / high-stakes
Deliberation

Judgment calls over your own codebase, docs, and history — where the reasons models disagree matter more than the average. The war room.

→ mumo
The loop

How a deliberation runs.

Inside the workflow you already have — your agent reaches for the panel, and you stay in the loop.

01

Your agent calls the panel with your context

Mid-task, on a call it shouldn't make alone, your agent invokes mumo over MCP — and brings your private context with it: the repo, the issue history, the design doc, your memories. The panel argues over your actual situation, not a generic one.

02

Models respond independently

No assigned roles, no model seeing another's answer first. Each writes its own raw position — so what surfaces is genuine independent reasoning, not a chorus shaped by whoever spoke first.

03

Models react to and comment on each other's responses

Now the deliberation: each model sees the others' positions and reacts to specific claims— championing, challenging, building on, or reframing them. Not parallel monologues — models engaging each other's strongest points.

04

Your agent gets the raw prose and the claim map

Back comes the full raw responses plus the claim map — a structured artifact that pulls every quote that drew a reaction and tags it by how each model responded: Keep, Challenge, Explore, Core, Shift. One glance shows where they converge and exactly where they split.

05

You — or your agent — make the call

Stop if you have what you need. Or append another round that automatically carries full context for every participant — feed in snippets and your own prose to steer. mumo never decides for you; the call stays with the party who knows your situation.

06

mumo keeps the full record

Every round is reviewable on a permanent page: the raw prose and the claim map, every claim traceable to the model that made it. Reasoning provenance — how the call was argued, on the record.

The claim map — what your agent gets back, alongside the raw prose

Five typed reactions turn a pile of opinions into something you can navigate — you see not just that the models disagreed, but how.

Keep

This holds up — endorse it and carry it forward.

Challenge

Push back, usually with a concrete alternative.

Explore

An open thread worth pursuing before you decide.

Core

The load-bearing point the rest hinges on.

Shift

Reframes the question or moves the angle.

The judge

A resident agent, not a tourist judge.

Every multi-model tool needs something to make sense of the panel. The question is who — a stranger who parachutes in once, or the agent that lives in your work.

The tourist judge

A stranger, once

A generic model that reads the panel once, writes the answer, and is gone.

  • ·Starts cold every time — no memory of the last call, or why you decided what you did.
  • ·Reads what's in the prompt and stops. It can't go pull the file it's missing and try again.
  • ·Writes the verdict and leaves. No next round, no steering, no follow-up.
Your resident agent

The one that's been here

The agent already in your harness — holding your context, your history, and the thread of how you decide.

  • +Remembers you rejected this approach last month, and challenges the panel when it forgets.
  • +Spots a gap, retrieves the file that settles it, and runs another round.
  • +Acts, not just summarizes — and has the history to act well.
Two forks of the same idea

A synthesizer returns the answer. mumo returns the map.

A synthesizer's single answer is a strong, fast tool — until the call is contested. Collapsing a panel into one answer can quietly bury the objection that should have changed your mind, and you can't see what it dropped. mumo takes the other fork: a structured, attributed map of where the models agreed, split, and why — the disagreement kept intact, not smoothed away. The answer still comes, just downstream — from the party grounded in your situation: you, or your agent.

Synthesis

A compound model

Panel in, judge compares, one model writes the answer.

  • +One clean call. Fast, low-effort, and cheap by comparison.
  • +Excellent for breadth-bound research with a verifiable best answer.
  • ·On a contested call, the collapse can bury the one objection that should have stopped you — and you can't see what it dropped.
  • ·A single pass — no rounds, no steering, no carrying context forward.
  • ·Per-call and ephemeral — nothing carries from one decision to the next.
A synthesizer is the right tool when you want an answer.
Deliberation

mumo

The same panel, but the disagreement is the deliverable you work with.

  • +A structured map, not a verdict — every reaction typed (Keep, Challenge, Explore, Core, Shift) and anchored to the exact claim it answers.
  • +Multi-round and steerable — context carries forward for every participant.
  • +Runs over your private context, pulled through your agent via MCP.
  • +It compounds — your agent carries how past calls went into the next one, and every round stays on the record.
Built for the call that's contested, situated, and worth working — when you need the judgment surface, not a verdict.
On the record

The reasoning behind the decision — kept.

mumo doesn't pretend to know what you did next, and it won't try to be your tracker. It keeps something narrower and truer: how the call was actually argued — which claims held up, what your context contributed, where the panel converged and where it didn't — reviewable whenever the decision comes back around.

Human as judge

Stop averaging your models. Make them argue over your actual work.

Connect your agent