metamedium
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFE
Full Analysis
- SAFE (SAFE): No malicious patterns, obfuscation, or exfiltration vectors were detected. The skill's instructions are limited to cognitive framing and standard user interaction.
- Indirect Prompt Injection (LOW): The skill analyzes user-provided descriptions of work (Phase 1) without explicit boundary markers or sanitization. However, its tool capabilities are limited to asking questions via 'AskUserQuestion', and it lacks access to sensitive systems or data, rendering the injection surface benign.
Audit Metadata