domains / families / software-engineering / error-debugging-multi-agent-review
Coordinate specialized code, security, architecture, and performance reviewers and consolidate their results. Use when a codebase needs parallel review across multiple disciplines. Review a defined code diff from relevant correctness, authorization and performance perspectives; delegate only when authorized and reproduce findings centrally.
The two copies describe the same task (multi-perspective code review) but disagree on the core procedure. The first copy is a high-level orchestration framework that mandates dynamic agent routing, parallel/sequential execution, weighted conflict resolution, and quality-score validation, and it includes Python pseudocode for an orchestration engine. The second copy is a bounded, safety-focused procedure that explicitly rejects those mechanisms (no routing code, no quality scores, no weighted votes), requires user authorization for delegation, and mandates central reproduction of findings. Because the first copy's core steps (routing, scoring, weighted voting) are explicitly forbidden by the second, crowning one would hide a real, conflicting alternative. The combined instruction must therefore present the two as distinct approaches and flag the conflicts.