nemotron-retrieval-recipes

Pass

Audited by Gen Agent Trust Hub on Jun 18, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill implements strict secret management guidelines, instructing agents to never ask users for API keys and to use environment variables or local configuration files instead.
  • [SAFE]: It enforces a 'dry-run first' policy, requiring agents to validate configurations with help commands and dry-run flags before initiating long-running, remote, or resource-intensive tasks.
  • [SAFE]: All remote resources, including datasets on Hugging Face and container images on the NVIDIA NGC registry, originate from well-known and trusted organizations.
  • [SAFE]: The skill includes explicit data governance warnings, advising users to check privacy policies before submitting proprietary data to synthetic data generation (SDG) endpoints.
  • [SAFE]: Instructions for remote execution via Slurm or other clusters include mandatory prerequisite checks and local validation steps to prevent accidental or unauthorized resource consumption.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 18, 2026, 04:34 PM
Security Audit — agent-trust-hub — nemotron-retrieval-recipes