clinical-decision-support
Pass
Audited by Gen Agent Trust Hub on Mar 31, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes Python scripts to perform statistical analysis and process clinical data inputs, which is consistent with its intended research purpose. It also generates and compiles LaTeX documents locally.
- [EXTERNAL_DOWNLOADS]: Documentation references the installation of standard, well-known data science libraries (pandas, numpy, scipy, lifelines, matplotlib) from public repositories via pip.
- [PROMPT_INJECTION]: Instructions in the skill definition use assertive language (e.g., 'MANDATORY', 'This is not optional') to guide the agent toward specific output formatting and structural requirements. These instructions do not attempt to circumvent agent safety protocols or safety filters.
- [SAFE]: The presence of a validation tool (validate_cds_document.py) that identifies potential HIPAA identifiers reflects a focus on data privacy and professional reporting standards.
Audit Metadata