Apex – AI-Powered Pentester Attacks Apps in Black-Box Mode to Find Vulnerabilities

Apex is an autonomous, AI-driven black-box penetration testing agent evaluated against Argus, an open-source benchmark of 60 Dockerized vulnerable web applications. The report describes Apex's deployment modes (CI, production, on-demand), testing results (271 unique vulnerabilities across multiple categories), comparative performance versus other tools, notable multi-step exploit successes, cost and failure modes, and availability as open source—focusing on tool capabilities and evaluation rather than any active breach or real-world incident.