midas_depth_estimation

Pass

Audited by Gen Agent Trust Hub on Mar 4, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill requires the installation of standard machine learning libraries including timm, torch, and torchvision from official package registries.
  • [REMOTE_CODE_EXECUTION]: The instructions involve loading a pre-trained model using torch.hub.load from the 'intel-isl/MiDaS' repository on GitHub. This operation fetches code and weights from the Intel Intelligent Systems Lab. As Intel is a well-known and trusted technology company, this is considered a safe and standard implementation for monocular depth estimation tasks.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 4, 2026, 03:12 AM