pydicom
Pass
Audited by Gen Agent Trust Hub on Jun 21, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides well-documented procedures for handling DICOM files using standard Python libraries. The included scripts (anonymize_dicom.py, dicom_to_image.py, extract_metadata.py) perform localized data processing tasks without any network access or suspicious external dependencies.
- [SAFE]: The anonymization script follows security best practices for medical data by providing a specific mechanism to strip Protected Health Information (PHI) from records.
- [SAFE]: No evidence of prompt injection, obfuscation, or persistence mechanisms was found in the documentation or the code. Instructions correctly advise handling exceptions when reading files from untrusted sources.
- [SAFE]: External dependencies are restricted to well-known and reputable packages in the Python ecosystem for scientific and medical data processing, such as pydicom, numpy, and pillow.
Audit Metadata