followup-tracking
Pass
Audited by Gen Agent Trust Hub on Sep 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes incidental medical findings provided by users, which serves as the ingestion point for untrusted data (SKILL.md). Boundary markers are absent in the prompt instructions. The capability inventory confirms that no subprocess calls, exec/eval, file-write, or network operations are requested or defined within the skill (SKILL.md). Sanitization of external input is not explicitly defined.
- [SAFE]: The skill uses non-sensitive placeholders for patient names, identifiers, and contact details in all examples and schemas, avoiding the hardcoding of actual PII.
Audit Metadata