Models Round Table

Answer one task with five models instead of one. Each model answers the raw prompt blind — you, the orchestrator, add no analysis, framing, or preferred answer before they respond, so the seats stay independent and unbiased. Then you reconcile: separate agreement from disagreement, run one disagreement round, bring in two judges, and make the final call. The deliverable is a single consensus answer plus a report of how it was reached.

Read-only: seats and judges answer and analyze; they do not modify the repo. (Turning a consensus into code is the job of feature-models-round-table / implement-and-review.)

Hard Rules

1. You are the moderator, not a seat. Add no analysis, hints, or preferred answer before the seats respond — the opening fan-out must be blind so none of your bias leaks in. Never count your own answer as a vote.
2. Read-only. Seats and judges produce answers/opinions only; they do not edit files, run mutating commands, or implement anything.
3. Blind opening. Every seat gets the raw prompt (plus read-only repo context only when the task is about this repo), no peer answers, no "right" answer.
4. Bounded reconciliation. Exactly one disagreement round, then a two-judge panel, then your final call. Do not loop indefinitely.
5. Never fabricate a seat or judge. Missing CLI/tool → drop it, lower confidence, and say so. Pass --disable-fallback to every runner so none silently borrows another provider.

Seats (5)

Launch each by its preferred path for the current host; fall back only when the native path is unavailable.