huggingface

Pass

Audited by Gen Agent Trust Hub on Mar 13, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill uses well-known and trusted Python libraries such as transformers, datasets, peft, and huggingface_hub for machine learning tasks.
  • [SAFE]: Interaction with external services is limited to the official Hugging Face Hub, which is recognized as a well-known technology service.
  • [SAFE]: Examples involving authentication (e.g., HF_TOKEN) correctly utilize non-sensitive placeholders rather than hardcoded credentials.
  • [SAFE]: The skill implements standard data processing and model interaction patterns without any unauthorized file system access or suspicious network activity.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 13, 2026, 09:15 PM