What problem does it solve? Single-reviewer code reviews can miss defects or produce unchallenged false positives. This Skill orchestrates an explicit adversarial review where two blind, read-only judges independently inspect the same immutable target, and only findings confirmed by both trigger bounded fixes. ## Core Features & Use Cases - Blind Dual Review: Launches two independent judges (jd-judge-a, jd-judge-b) in parallel against one frozen target with identical scope and criteria, then merges their findings into a frozen ledger. - Bounded Fix Rounds: Applies fixes only to severe findings confirmed by both judges, with at most two fix rounds and two scoped re-judgments that inspect only the ledger plus fix delta. - Deterministic Verdicts: Returns exactly one terminal verdict, JUDGMENT: APPROVED or JUDGMENT: ESCALATED, with confirmed/suspect/contradiction counts and artifact references. - Use Case: Before merging a critical authentication module, request a Judgment Day review so two independent judges verify correctness, security, and error handling, with contradictions escalated for human decision. ## Quick Start Ask the assistant to run Judgment Day dual review on the specific files or module you want adversarially reviewed.
