training-loop

Pass

Audited by Gen Agent Trust Hub on Jun 19, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill outlines a logical sequence of machine learning tasks using standard command-line tools such as yolo-train and yolo-validate.
  • [SAFE]: No hardcoded credentials, unauthorized network requests, or attempts to download external scripts were detected.
  • [SAFE]: The skill ingests a local configuration file (yolo-project.yaml) to manage training parameters, which is an expected behavior for this use case and does not indicate an indirect prompt injection risk.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 19, 2026, 02:47 PM
Security Audit — agent-trust-hub — training-loop