What problem does it solve? Security scans produce candidate findings, but deciding whether each one is a real, reportable vulnerability requires tracing attacker reachability, trust boundaries, and impact from repository evidence. This Skill turns validated findings into explicit attack-path analyses with calibrated severity and a final reportability decision. ## Core Features & Use Cases - Attack Path Construction: Builds factual attacker stories from repository evidence covering exposure, entry points, identities, trust boundaries, secrets handling, and reachability. - Counterevidence Analysis: Identifies the strongest repository evidence against scoping and reportability fields before finalizing decisions. - Severity Calibration and Policy Adjustment: Applies a mechanical impact/likelihood matrix and suppression rules to produce final severity, priority (P0-P3), and ignore/report decisions. - Use Case: During a repository security scan, a candidate SQL injection finding is validated; this Skill traces whether attacker input actually reaches the sink from an in-scope surface, weighs counterevidence, and decides whether it survives as a reportable high-severity finding. ## Quick Start Ask the agent to run attack-path analysis on the validated findings from the current security scan and produce severity-calibrated reportability decisions.
