What problem does it solve? After implementing a plan, code often drifts from the original intent, introduces security or reliability risks, or breaks existing conventions. This Skill compares actual implementation work against the original plan to catch drift, dangerous decisions, architecture violations, and pattern misuse before they compound. ## Core Features & Use Cases - Plan Drift Detection: Uses parallel sub-agents to verify each planned change matches the actual implementation, flagging missing items, intent mismatches, and unplanned scope creep. - Safety & Quality Scan: Checks changed files for security risks (injection, hardcoded secrets), performance issues (N+1 queries), reliability gaps, and data safety problems, plus pattern consistency against sibling files. - Structured Report & Interactive Triage: Produces a severity-ranked findings report (CRITICAL/WARNING/OBSERVATION) with dimension verdicts and an overall APPROVED/NEEDS ATTENTION/REJECTED verdict, then walks each finding through an interactive fix/skip/record-as-lesson triage loop. - Use Case: After finishing phase 3 of a multi-phase feature plan, run a phase-scoped review to confirm the implementation matches the plan, automated success criteria pass, and no unplanned endpoints were added before moving on. ## Quick Start Ask the AI to review the implementation against the plan at context/changes/my-change/plan.md and triage any findings.
