propensity-score-matching

Pass

Audited by Gen Agent Trust Hub on Sep 8, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [SAFE]: The skill is a legitimate tool for statistical analysis, providing a complete workflow for propensity score matching. It operates entirely on local data using standard Python mathematical libraries.- [COMMAND_EXECUTION]: The scripts/validate_fixture.py script uses subprocess.run to execute the analysis engine for validation. This is a standard testing practice and does not process untrusted external commands.- [DATA_EXPOSURE]: The skill documentation includes explicit safety requirements for data handling, advising against the use of authentication secrets and emphasizing read-only interactions with data platforms.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 8, 2026, 03:23 AM
Security Audit — agent-trust-hub — propensity-score-matching