methodology

A phase-by-phase checklist for pentesting an LLM app — recon, prompt injection, jailbreaks, RAG abuse, and access control. Distilled from my AI Hacking 101 notes.