Search Results for "attacks"

...The workflow begins with reconnaissance, then moves into parallel hunting across attack classes such as injection, access control, business logic, cryptography, feature abuse, and chained attacks. Findings are then challenged through separate validation agents to reduce false positives. The skill produces human-readable reports, detailed finding traces, structured JSON output, and independent verification against the actual source code. Overall, it is designed to make AI-assisted security review more systematic, adversarial, and evidence-based.