mama
Pass
Audited by Gen Agent Trust Hub on Mar 27, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill primarily focuses on narrative role-play and persona-based task motivation. No malicious patterns were identified across the 10 threat categories.
- [PROMPT_INJECTION]: The skill uses instructions to adopt a specific persona and behavioral traits (high persistence), but these are framed as task-oriented motivation rather than attempts to bypass system safety guidelines or ethical constraints.
- [DATA_EXFILTRATION]: The skill instructs the agent to locate and read an internal skill definition file (
**/pua/skills/pua/SKILL.md) to inherit its behavior. This is a localized file operation within the skill's own package structure and does not target sensitive user data or involve external network communications.
Audit Metadata