huggingface-accelerate

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADS
Full Analysis
  • [SAFE] (SAFE): No malicious patterns detected. The skill contains instructional content and code snippets for machine learning developers using industry-standard tools.
  • [EXTERNAL_DOWNLOADS] (LOW): The skill references the installation of common Python packages (accelerate, torch, transformers, deepspeed) from PyPI. These are trusted libraries within the machine learning ecosystem.
  • [COMMAND_EXECUTION] (LOW): The documentation includes standard CLI commands for the accelerate tool, which are necessary for its primary purpose of launching distributed training jobs.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:25 PM