Skills
Navigate
skills/drshailesh88/integrated_content_os/clinical-decision-support/Gen Agent Trust Hub

clinical-decision-support

Pass

Audited by Gen Agent Trust Hub on Nov 6, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted patient data from external CSV and JSON files to perform biomarker stratification and clinical reporting.
  • Ingestion points: Data entering the agent context via scripts like biomarker_classifier.py and create_cohort_tables.py.
  • Boundary markers: No explicit delimiters or instructions are used to distinguish processed data from the agent's instructions.
  • Capability inventory: The skill utilizes Read, Write, Edit, and Bash tools.
  • Sanitization: The provided validation script focuses on HIPAA compliance rather than sanitizing against adversarial prompt instructions.
  • [EXTERNAL_DOWNLOADS]: The documentation specifies dependencies on several well-known Python packages for data analysis.
  • Evidence: README.md and SKILL.md list pandas, numpy, scipy, lifelines, matplotlib, and pyyaml.
  • Note: These are trusted libraries commonly used in scientific research.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 6, 2026, 12:15 PM