Results for “adversarial-attacks”

Audits GitHub Actions workflows for security vulnerabilities in AI agent integrations, detecting attack vectors where attacker-controlled input reaches AI agents in CI/CD pipelines.

Reverse-engineers EDR, Defender, and AV hook tables, ETW providers, and AMSI implementations to build targeted bypasses including unhooking, indirect syscalls, ETW patching, and call stack spoofing for authorized red team operations.