shap-model-explainability

Pass

Audited by Gen Agent Trust Hub on May 6, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill serves as a technical guide for the SHAP library, a widely used tool for model explainability. No malicious behavior, prompt injections, or data exfiltration attempts were detected.
  • [EXTERNAL_DOWNLOADS]: The skill provides instructions to install standard machine learning packages (shap, matplotlib) from public registries and includes code to download example datasets through official library functions (e.g., shap.datasets.adult()).
  • [COMMAND_EXECUTION]: Mentions common shell commands for environment setup, such as pip install, which is expected for a development-oriented skill.
Audit Metadata
Risk Level
SAFE
Analyzed
May 6, 2026, 01:45 AM
Security Audit — agent-trust-hub — shap-model-explainability