Reviewer Pattern — Problem

Pattern: Reviewer
Component: problem.md
Version: 1.1 | Updated: 2026-07-16

---

Statement

Pull requests and agent-generated diffs merge with consequential defects because review focuses on style, trusts green CI, or skips requirement tracing and trust-boundary walks. The result is production incidents that were visible in the diff.

Measurable symptoms

Symptom	How you detect it
Escaped defects	Post-merge bugs that a severity-calibrated review would have blocked
Nit flooding	>15 Minor comments and zero Critical/High on a behavioral PR
Untested AC	Acceptance criteria with no corresponding test in the PR
Self-approve	Same agent/human authored and gave final Approve on High-risk change
Unverified “fixed”	Finding marked resolved without re-check on the new revision

Root cause

Reviewers optimize for speed or comment volume. Defect detection requires reconstructing intended behavior, walking trust boundaries, and calibrating severity from reachability × impact — work that is skipped under time pressure or when CI is treated as proof.

Non-goals of this pattern

• Not a substitute for automated SAST/DAST (complements them)
• Not a place to rewrite the feature for the author
• Not required for docs-only or pure formatting commits
• Not an architecture redesign session (escalate to architecture-pattern)

Related failure modes

See failures.md for recovery when reviews are nit-heavy, CI-blind, self-approved, or severity-miscalibrated.