train-test-split

Pass

Audited by Gen Agent Trust Hub on Jul 7, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill contains no executable code, scripts, or subprocess commands.
  • [SAFE]: No network operations, external downloads, or remote resource references were found.
  • [SAFE]: No prompt injection, obfuscation, or data exfiltration patterns were detected in the instructions or test cases.
  • [SAFE]: The methodology described (training/test split) is a standard industry best practice for machine learning.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 7, 2026, 04:04 AM
Security Audit — agent-trust-hub — train-test-split