pyhealth
Pass
Audited by Gen Agent Trust Hub on Jun 26, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides legitimate documentation and code examples for the PyHealth healthcare AI library. The content is educational and instructional, focusing on clinical machine learning workflows.
- [EXTERNAL_DOWNLOADS]: The skill references the installation of the 'pyhealth' package from standard Python package registries using the 'uv' or 'pip' package managers. This is a standard and safe dependency management practice.
- [DATA_EXPOSURE]: Analysis of data handling patterns shows no unauthorized access to sensitive system files or hardcoded credentials. The documentation uses standard placeholders for data paths, and the library's data structures (Patient, Visit, Event) are designed for local processing of research datasets.
- [COMMAND_EXECUTION]: The provided code examples use standard Python libraries (NumPy, pandas, PyTorch) for data science tasks. No arbitrary shell command execution, privilege escalation attempts, or persistence mechanisms were found.
- [PROMPT_INJECTION]: The instructions do not contain any patterns attempting to override agent safety guidelines, extract system prompts, or bypass behavioral constraints.
Audit Metadata