What problem does it solve? Single-reviewer code reviews miss blind spots and reflect one model's biases. This Skill runs the same change past several independent LLM reviewers and distills their findings into one prioritized, actionable verdict. ## Core Features & Use Cases - Multi-Model Adversarial Review: Spawns one reviewer per configured model, each applying the same prompt, rubric, and code-quality lens so the signal comes from model diversity. - Lead Judgment Synthesis: Deduplicates findings, flags consensus issues raised by 2+ models, and categorizes every finding as act on, consider, noted, or dismissed. - Configurable Reviewer Roster: Reads reviewer model assignments from a rules file with sensible defaults and graceful fallback when a model slug is unresolvable. - Use Case: Before merging a feature branch, ask for an adversarial review of the diff against main and receive a structured verdict listing what must be fixed, what to consider, and what was dismissed with rationale. ## Quick Start Ask the assistant to interrogate the current branch diff against main and produce a synthesized multi-model review verdict.
