NYC

ml-pipeline-workflow

Pass

Audited by Gen Agent Trust Hub on Feb 16, 2026

Risk Level: LOW
Full Analysis
  • General Security (SAFE): The skill provides high-level guidance, architecture patterns, and templates for MLOps workflows. It does not include any functional code or scripts that could be executed maliciously.- Indirect Prompt Injection (LOW): The skill outlines workflows for ingesting raw data from external sources (SKILL.md). Evidence: 1. Ingestion points: Raw data sources. 2. Boundary markers: Absent in documentation. 3. Capability inventory: None (informational only). 4. Sanitization: Advocates for Great Expectations and TFX (SKILL.md). The risk is minimal as the skill is primarily guidance.- Data Exposure & Exfiltration (SAFE): The skill references cloud platforms like AWS SageMaker and Google Vertex AI but does not contain or request any hardcoded credentials, API keys, or sensitive configuration details.
Audit Metadata
Risk Level
LOW
Analyzed
Feb 16, 2026, 10:56 AM