What problem does it solve? Verifying that implemented code actually matches the product specification is tedious and error-prone; this Skill audits completeness, quality, security, and type safety against Product-Spec.md and design mockups, producing a structured report with file:line evidence for every conclusion. ## Core Features & Use Cases - Spec Compliance Audit: Checks every functional requirement in Product-Spec.md item by item, classifying each as fully, partially, or not implemented, and flags Spec Drift for out-of-scope code. - Multi-Dimension Parallel Review: For moderate or complex changes, dispatches four review dimensions (design, bug, security, types) whose findings are aggregated with confidence scoring and risk ranking. - Evidence-Based Reporting: Every finding includes severity, impact, confidence, risk_rank, and file:line evidence, sorted into Must-fix, Should-fix, and Insight buckets with a final ship/fix-first/blocked verdict. - Use Case: After a development phase completes, run the review to confirm all Spec features exist, catch hardcoded credentials or any type usage, and receive a prioritized fix list before sign-off. ## Quick Start Ask the AI to run /code-review to audit the current codebase against Product-Spec.md and produce an aggregated review report.
