degenerate-input-filtering

Pass

Audited by Gen Agent Trust Hub on May 6, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides documentation and reference code for data preprocessing tasks, specifically identifying and removing uninformative data features (constant values, all-NaN columns) prior to statistical testing to prevent silent failures and NaN propagation.
  • [SAFE]: Code examples utilize standard, well-known libraries (pandas, numpy, scipy) for local data manipulation and verification.
  • [SAFE]: External links point to official documentation (SciPy, Pandas, NumPy) and reputable academic DOI repositories, which are considered safe sources.
  • [SAFE]: No network operations, sensitive file access, obfuscation, or arbitrary command execution patterns were detected.
Audit Metadata
Risk Level
SAFE
Analyzed
May 6, 2026, 01:47 AM
Security Audit — agent-trust-hub — degenerate-input-filtering