Insecure output handling: LLM response to stored XSS
When an app renders model output as HTML without sanitizing, an attacker who controls part of the prompt gets XSS in the victim's session.
When an app renders model output as HTML without sanitizing, an attacker who controls part of the prompt gets XSS in the victim's session.
A polluted prototype property read by a sink like a template loader or script src becomes reliable XSS without any server bug.