edc-data-validation

Pass

Audited by Gen Agent Trust Hub on Jul 9, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill generates deterministic Python code for clinical data validation tasks. All logic is focused on data manipulation, range checking, and file generation (CSV and XML) using standard, well-known libraries.
  • [COMMAND_EXECUTION]: The generated code is intended for local execution by the user for data science workflows. It does not use unsafe functions like eval() or os.system(), nor does it attempt to access sensitive system files or credentials.
  • [DATA_EXFILTRATION]: No network operations (e.g., requests, curl, socket) were detected. Data processing is performed on locally provided DataFrames, and output is saved to local paths specified by the user.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 9, 2026, 05:59 PM
Security Audit — agent-trust-hub — edc-data-validation