What problem does it solve? Code reviews often miss non-obvious defects like race conditions, silent error swallowing, unsafe type narrowing, and AI-generated low-quality patterns that pass type-checking but fail in production. This Skill performs adversarial static analysis to surface these hidden issues before they cause outages. ## Core Features & Use Cases - Adversarial Static Analysis: Scans files, folders, or entire repositories for 13 bug categories including async races, security flaws, memory leaks, and state corruption. - Structured Severity Model: Every finding is rated by severity, confidence, and blast radius, then delivered in a strict report with an executive summary, findings table, deep dives, and an action plan. - AI Slop Detection: Identifies vague helper names, lazy abstractions, copy-paste patterns, and defensive code without a threat model. - Use Case: Point it at a TypeScript service directory before a release to catch incorrect memoization, improper cache invalidation, and partial failure scenarios that unit tests missed. ## Quick Start Ask the assistant to run a bug-hunter scan on a specific file, folder, or the entire repository and return the prioritized findings report.
