Ultra mode

The base pipeline plans with a single genius. The ultra variants (/ultraship, /ultrashipit) run the exact same pipeline but replace that one planner with two genius seats that debate the plan before anything builds, then hand the agreed plan to the unchanged execute-and-consolidate machinery. Ultra reviews genius-tier too.

The two seats

Both are model-role seats you configure in modelRoles.

• plato is the chief architect and final consolidator. It always owns THE plan.
• aristotle is the challenger. It red-teams and never rubber-stamps. It's better to use a different model for this.

Why the pipeline resolves the chains itself

On omp ≤ 16.3.15 the pi/<role> alias resolver was hard-gated to built-in role names: a custom role like pi/plato passed through unresolved, and omp then silently spawned an undefined-model session, discarding the error. omp ≥ 17 lifts that gate — a custom @plato resolves for any configured modelRoles key.

The pipeline still does not rely on the alias. Cell 1 reads modelRoles from config itself for two reasons: an unconfigured @plato resolves to nothing without an error at the call site, whereas the config read asserts both seats loudly; and a self-resolved comma-joined chain passed via model= behaves identically on every omp version. A call-site model= overrides the planner agent's own frontmatter chain, so the geniuses you configured are the ones that run. (Note: eval's completion() helper is separate — its model param is a closed smol|default|slow enum, so custom roles never work there.)

Plan-time freezing

The two resolved chains are frozen into the run state at plan time, stored in meta.ultra alongside the topology. Editing modelRoles mid-run does not retarget an in-flight or resumed run. The review stage reads the seats from that frozen state, not live config, so a resumed run keeps its ultra identity and reviews with the same brains that planned. Plan-time identity is deliberate.

Loud-fail

Ultra refuses to degrade to a single genius. Cell 1 asserts that both plato and aristotle are configured and non-empty. If either is unset, the run fails loud rather than quietly planning with one seat.

Fresh-eyes consult

In an ultra run, a design escalation from a stuck builder is re-adjudicated by the challenger (aristotle), not the plan's author. The model that red-teamed the plan is best placed to reopen it. bug escalations still go to deep-debugger.

Where to go next

• Planning topologies covers crossreview, duel, and debate.
• Ultra review covers the genius review duel.