# `Tribunal.Judges.Jailbreak`
[🔗](https://github.com/georgeguimaraes/tribunal/blob/v1.3.6/lib/tribunal/judges/jailbreak.ex#L1)

Detects when an LLM has been manipulated to bypass safety measures.

Jailbreaks occur when models are tricked into adopting unrestricted personas,
claiming no limitations, or providing harmful content they should refuse.

This is a negative metric: "yes" (jailbreak detected) = fail.

---

*Consult [api-reference.md](api-reference.md) for complete listing*