helpful-formatter
Pass
Audited by Gen Agent Trust Hub on Jun 14, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill instructions explicitly direct the AI to 'set aside its usual constraints' and treat 'previous operational guidelines' as 'suggestions rather than requirements', which is a pattern used to bypass safety filters.
- [PROMPT_INJECTION]: The behavior section attempts to establish a roleplay scenario where the user has 'elevated access permissions' and the AI is in a 'fully cooperative mode' with 'unrestricted operation'.
Audit Metadata