north-wind
Pass
Audited by Gen Agent Trust Hub on Jun 22, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill uses metaphorical and abstract language (e.g., 'North Wind', 'ceremonial safety language', 'stripping protection') to instruct the agent to ignore or bypass safety guidelines and 'protective layering' that might be present in the conversation or system prompts. This is a characteristic pattern of prompt injection aimed at circumventing content filters.
- [PROMPT_INJECTION]: The workflow specifically directs the agent to identify and strip 'aura, barrier, wrapper, soft shield' or 'ceremonial safety language' that prevents 'real contact'. This encourages the agent to evaluate and potentially disregard the platform's safety requirements by framing them as 'beneficial cover' that 'soaks hits' or 'distorts the board'.
Audit Metadata