Jailbreak Classifier

Scores an inbound prompt for jailbreak/persona-override risk before the agent acts on it.

gl0di Updated

File contents

Jailbreak Classifier

Scores an inbound prompt and returns a risk band so the calling agent can decide whether to escalate to a human.

Usage

python scripts/score_prompt.py < prompt.txt

gl0di/clawseccheck/tree/main/fixtures/clean_b66_script_docstring_classifier/skills/jailbreak-classifier commit da1b16b50b

Frequently asked questions

npx skillmds@latest add gl0di/jailbreak-classifier