vendor-due-diligence

Pass

Audited by Gen Agent Trust Hub on Oct 30, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides two Python scripts, scripts/vendor_risk_scorer.py and scripts/vendor_comparison.py, which are used to automate risk scoring based on vendor questionnaires. A detailed review of the source code confirms that these scripts only use standard Python libraries (json, argparse, sys) to perform data processing. They do not perform network requests, file-system modifications, or any form of dynamic code execution. The instructions are clearly written and do not contain any patterns indicative of prompt injection or obfuscation. The skill's architecture relies on local JSON files for input and produces standard output for the user, aligning with secure development practices for AI agent extensions.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 30, 2026, 10:33 AM