dicom-metadata-extractor
Pass
Audited by Gen Agent Trust Hub on Sep 21, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process external medical imaging data (DICOM files), which represents a potential surface for indirect prompt injection if metadata fields contain adversarial instructions. 1. Ingestion points: DICOM medical images located in user-specified directories (e.g., /data/imaging/). 2. Boundary markers: None explicitly defined to separate untrusted metadata content from agent instructions. 3. Capability inventory: File system read/write, batch metadata extraction, and report generation (CSV/JSON). 4. Sanitization: The skill implements robust PHI identification and redaction following HIPAA Safe Harbor guidelines, which acts as a security control for data privacy.
Audit Metadata