deep-understanding
Pass
Audited by Gen Agent Trust Hub on Jun 16, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No security issues detected. The skill serves as a high-level router and instructional guide for educational purposes.
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided technical content (papers, methods, or algorithms). While this presents a theoretical surface for indirect prompt injection, the skill acts solely as an orchestrator and does not exhibit capabilities for file system modification, network exfiltration, or code execution that could be exploited via this vector.
Audit Metadata