What problem does it solve? Code reviews often miss security regressions hidden in diffs, especially when validation checks are removed or previously fixed vulnerabilities are reintroduced. This Skill systematically analyzes PRs, commits, and diffs to detect security regressions, missing test coverage, and high blast radius changes before they reach production. ## Core Features & Use Cases - Risk-Adaptive Analysis: Classifies codebases as SMALL, MEDIUM, or LARGE and scales review depth accordingly, prioritizing HIGH RISK changes like auth, crypto, and value transfer logic. - Git History Forensics: Uses git blame and commit history to detect removed security checks, re-added vulnerable code, and regressions of previous CVE fixes. - Adversarial Modeling: Builds concrete attacker models and step-by-step exploit scenarios with exploitability ratings for HIGH RISK changes. - Use Case: A team submits a PR modifying a withdraw function in a smart contract. The Skill detects a removed require check, traces it back to a prior security fix commit, calculates that 23 callers are affected, and generates a markdown report with a CRITICAL finding and proof-of-concept exploit. ## Quick Start Ask the AI to perform a differential security review of the changes between two commits or a pull request and generate a full markdown report.
