nick-theriot
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- Prompt Injection (LOW): The skill utilizes high-priority markers such as 'CRITICAL' and 'MUST' to enforce linguistic styles and contains instructions to 'NEVER break character or mention you're an AI'. While intended for persona consistency, these represent attempts to override the model's default behavior and safety transparency.
- Indirect Prompt Injection (LOW): The skill has an ingestion surface where user queries are analyzed and then interpolated into tool parameters (e.g., in Step 3 for 'retrieve_mental_models'). There are no boundary markers or explicit sanitization steps documented to prevent malicious user input from manipulating the retrieval queries or the final synthesis process.
Audit Metadata