medical-device-software

Pass

Audited by Gen Agent Trust Hub on Nov 11, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill uses four Python scripts (software_safety_classifier.py, lifecycle_gap_check.py, soup_register_check.py, and trace_matrix_check.py) that are implemented exclusively using the Python standard library. No external packages are installed or requested.
  • [SAFE]: No network communication, subprocess execution, or data exfiltration patterns were detected in any of the scripts or instructions.
  • [SAFE]: Reference links point to authoritative regulatory and technical documentation from sources such as the FDA, European Commission, and established industry standards bodies.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests user-provided data via CSV and JSON files for compliance checking. 1. Ingestion points: software_safety_classifier.py (--inventory), lifecycle_gap_check.py (--assessment), soup_register_check.py (--register, --sbom), and trace_matrix_check.py (--matrix). 2. Boundary markers: Absent. 3. Capability inventory: No subprocess calls, exec/eval, file-writing, or network operations in software_safety_classifier.py, lifecycle_gap_check.py, soup_register_check.py, or trace_matrix_check.py. 4. Sanitization: Data is processed using standard CSV and JSON parsers with no execution of content. (Severity: SAFE)
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 11, 2026, 01:04 PM