pyhealth
Pass
Audited by Gen Agent Trust Hub on Feb 15, 2026
Risk Level: LOW
Full Analysis
- [SAFE] (SAFE): The skill consists of informational documentation for a legitimate medical research library. No executable malicious code, obfuscation, or command execution patterns were found.
- [INDIRECT_PROMPT_INJECTION] (LOW): While the skill describes tools for ingesting clinical notes and EHR data—which are potential surfaces for indirect prompt injection—no exploitable logic or autonomous decision-making capabilities are defined in the analyzed files.
- [DATA_EXFILTRATION] (SAFE): No hardcoded credentials or unauthorized network operations were identified; the examples use local paths for data loading.
Audit Metadata