ml-data-pipeline

Pass

Audited by Gen Agent Trust Hub on Jun 13, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill serves as a comprehensive guide for ML data engineering workflows, providing instructions for data ingestion, validation, and feature engineering.
  • [EXTERNAL_DOWNLOADS]: The skill references official documentation for several well-known technology projects including TensorFlow (Google), Apache Spark, Feast, DVC, and Polars. These references are used for educational purposes and point to established, official domains.
  • [COMMAND_EXECUTION]: Provides standard command-line examples for Data Version Control (DVC). These commands are routine for managing data versions and do not contain malicious payloads or privilege escalation attempts.
  • [SAFE]: Code snippets for processing engines (Spark, Polars) and feature stores (Feast) follow standard usage patterns and do not involve unsafe dynamic code execution or sensitive data exfiltration.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 13, 2026, 03:39 AM
Security Audit — agent-trust-hub — ml-data-pipeline