What problem does it solve? Single-reviewer code reviews carry one model's blind spots and biases, making it easy to miss bugs, design flaws, and maintainability risks in a changeset. ## Core Features & Use Cases - Multi-Model Adversarial Review: Spawns one reviewer per configured model so independent models challenge the same diff from different angles, turning cross-model agreement into high-confidence signal. - Structured Rubric and Quality Lens: Each reviewer applies a shared rubric covering correctness, root causes, structural integrity, verification, complexity, and security, plus a strict code-quality lens targeting structural simplification. - Lead Judgment Synthesis: A lead-reviewer step filters findings into Act On, Consider, Noted, and Dismissed buckets, deduplicates across models, and produces an agreement map. - Use Case: Before merging a feature branch, run an adversarial review on the full diff so several models independently hunt for bugs and design flaws, then receive a single prioritized verdict instead of four raw review dumps. ## Quick Start Ask the agent to interrogate the current branch changes against main and produce a synthesized adversarial review verdict.
