health-data-model-planning
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill recommends the use of established libraries from trusted organizations and the vendor's own ecosystem. These include the apple/FHIRModels repository and the stanfordspezi project's GitHub and NPM resources. These references are used for standard implementation guidance.\n- [INDIRECT_PROMPT_INJECTION]: The skill represents a surface for indirect prompt injection as it processes untrusted user data to generate documentation.\n
- Ingestion points: User-provided domain information, health concepts, and product requirements described during the planning phase.\n
- Boundary markers: None identified in the prompt instructions to delimit user-provided content.\n
- Capability inventory: The skill generates a planning brief intended to be written to
docs/planning/data-model-brief.mdin the local repository.\n - Sanitization: No explicit sanitization or validation of the input data is performed before interpolation into the planning brief.
Audit Metadata