medic
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted medical data from various external sources, creating a surface for potential indirect prompt injection attacks.\n
- Ingestion points: Processes medical PDFs, free-text notes, and FHIR data bundles as described in
SKILL.md.\n - Boundary markers: No specific delimiters or 'ignore instructions' warnings are defined for the input data.\n
- Capability inventory: The skill contains no file-system write operations, network requests, or command execution capabilities.\n
- Sanitization: The instructions include terminology normalization and illegibility flagging, but no specific sanitization for adversarial text patterns.
Audit Metadata