computer-vision-guide

Pass

Audited by Gen Agent Trust Hub on Mar 31, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill serves as a technical reference for computer vision tasks and research methodology.
  • [COMMAND_EXECUTION]: The provided Python code snippets use standard libraries such as os and pathlib for organizing local image datasets. These operations are benign and consistent with the skill's stated purpose of dataset preparation.
  • [EXTERNAL_DOWNLOADS]: The skill includes code for initializing models with pre-trained weights via torchvision.models. These weights are retrieved from official, well-known PyTorch repositories during runtime, which is a standard and safe practice in machine learning development.
  • [PROMPT_INJECTION]: No instructions were found that attempt to override agent behavior, bypass safety filters, or extract system prompts.
  • [DATA_EXFILTRATION]: No network operations, hardcoded credentials, or sensitive file access patterns were detected.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 31, 2026, 10:16 PM