What problem does it solve? Reviewing pull requests and branch changes manually is slow and inconsistent, often missing security flaws, language-specific anti-patterns, and performance regressions. This Skill runs a structured, repeatable review of any branch diff so nothing slips through before merge. ## Core Features & Use Cases - Multi-language strictness checks: Auto-detects TypeScript, Rust, Swift, Go, and JavaScript files in the diff and applies the relevant rule sets (e.g., no any, no .unwrap(), no force unwrap). - Security and performance scanning: Flags hardcoded secrets, injection risks, missing auth checks, O(n^2) loops, sync I/O in async contexts, and unbounded caches. - Severity-ranked report: Outputs findings grouped as CRITICAL, HIGH, MEDIUM, and LOW with file, line, and rule ID, plus a passed-checks summary. - Use Case: Before merging a feature branch, run the review against main to catch a leaked API key, an empty catch block, and a nested-loop performance issue in one pass. ## Quick Start Ask the assistant to run /code-review on the current branch to get a severity-ranked review of all changes against main.