council
Parallel multi-model consensus review with configurable perspectives, debate rounds, and deep exploration.
- Supports three task modes: validate (PASS/WARN/FAIL verdicts), brainstorm (explore alternatives), and research (deep analysis with trade-offs)
- Spawns 2–3 independent judges by default; scales to 6+ agents with
--deep,--mixed(Claude + Codex), and--explorerssub-agents, capped at 12 total - Debate mode (
--debate) runs two-round adversarial review with verdict exchange and position revision for high-stakes decisions - Built-in presets (security-audit, architecture, code-review, plan-review, doc-review, retrospective) assign perspective-specific prompts; custom perspectives via
--perspectivesor YAML file
/council — Multi-Model Consensus Council
Convene N independent reasoners over a shared briefing and return one synthesis. --mode selects the deliberation pattern — brainstorm, debate, or verdict. Everything else (--focus, --depth, --runtime, --roster) is an orthogonal knob.
Loop position
Cross-cutting judgment gate available at any operating loop move where multi-model consensus is required: typically pre-flight on the slice plan (between moves 3 and 4), per-slice on non-mechanical correctness (move 6), and on the bead acceptance roll-up. Council does not own a loop move — it provides verdicts that other moves consume. Use it at slice level when a single test cannot capture taste; use it at bead level when acceptance examples are passing but the consumer-facing behavior still needs adversarial review.