writing-skills
Fail
Audited by Gen Agent Trust Hub on Feb 16, 2026
Risk Level: HIGHPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION] (HIGH): The skill provides explicit instructions on using 'Authority' and 'Non-negotiable framing' to eliminate AI rationalization. This approach is designed to override safety filters and ethical reasoning by using imperative commands.
- [PROMPT_INJECTION] (HIGH): The document references specialized techniques for 'Persuading AI to Comply with Objectionable Requests', effectively serving as a guide for high-success rate prompt injection and safety bypass.
- [PROMPT_INJECTION] (MEDIUM): It outlines how to use 'Commitment' and 'Social Proof' to lock the AI into a behavioral loop that favors compliance over safety guidelines, mirroring patterns found in sophisticated jailbreaking attempts.
Recommendations
- AI detected serious security threats
Audit Metadata