NYC

clinical-reports

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFE
Full Analysis
  • [PROMPT_INJECTION] (SAFE): No patterns of instruction override, jailbreaking, or system prompt extraction were found in the skill's documentation or templates.
  • [DATA_EXPOSURE] (SAFE): The skill does not access sensitive system paths or hardcoded credentials. It explicitly includes tools for HIPAA compliance and de-identification of patient data.
  • [REMOTE_CODE_EXECUTION] (SAFE): There are no network requests, remote script downloads, or dynamic execution patterns (eval/exec) in the provided Python scripts.
  • [COMMAND_EXECUTION] (SAFE): Local file operations are limited to standard file reading and template copying via the shutil library. No arbitrary shell command execution was detected.
  • [EXTERNAL_DOWNLOADS] (SAFE): No external dependencies or package installations are required beyond the Python standard library.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:27 PM