risk-management-specialist
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill provides a local Python script for risk matrix and FMEA calculations using standard library modules with no network or sensitive file access. All instructional content aligns with regulatory compliance best practices.
- [INDIRECT_PROMPT_INJECTION]: The skill ingests user-provided device descriptions in SKILL.md to perform risk analysis. It implements boundary markers by instructing the agent to list assumptions. The skill's capabilities are restricted to text generation and a local calculation utility, with no access to external networks or privileged system commands, and it lacks explicit sanitization for user-provided clinical context.
Audit Metadata