writing-skills
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
- Prompt Injection (LOW): The guide explicitly advocates for the use of authority-based imperative language such as 'YOU MUST', 'Never', 'Always', and 'No exceptions' to ensure compliance and eliminate rationalization. These markers are commonly used in adversarial prompt injections to override system instructions and safety guardrails.
- Prompt Injection (LOW): The document describes using 'Commitment' and 'Scarcity' tactics (e.g., 'IMMEDIATELY') to force automatic behavior and reduce cognitive load on compliance. These techniques can be used to pressure an AI model into executing instructions before safety filters or ethical reasoning can be fully applied.
Audit Metadata