parallel-code-review

Code review when parallel diverse lenses catch issues single reviewer misses

Autonomous merge gates for SceneView PRs: reviewers catch issues, adversarial verify filters false positives, breaking-API change blocks auto-merge with traceback.

Blocking unsafe SceneView merges with enforced multi-lens coverage (correctness, API, parity, UX, safety) and a serial visual QA gate that prevents device-contention race conditions.

Multi-dimensional acceptance criteria requiring independent per-AC verification.

Secondary review gate: catch sycophancy, false positives, and missed issues in bot reviews; converge on approve/request-changes/defer via up to 10 rounds.

Auditing code quality before merge, catching cross-cutting issues (errors, tests, types) that single reviewers miss

Systematic PR review before merge or to supplement human review

Structured four-pass review (security/performance/correctness/maintainability) with optional GitHub posting

Parallel specialist reviews (security, perf, correctness) in one coordinated workflow