» Tag
pentesting
2 posts6–Jul 12, 2026
Building an AI that pentests your AI, with proof required
An engineer built an adversarial LLM harness that only reports proven exploits against a live agent, replacing vague 'might be vulnerable' pentest reports with deterministic, oracle-verified findings.
Windows Privilege Abuse: The Shortcut to SYSTEM Access
Exploring how Windows privileges like SeImpersonatePrivilege are weaponized into SYSTEM access via Potato attacks, and why architectural defenses matter.