flide
Pass
Audited by Gen Agent Trust Hub on Nov 13, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill demonstrates strong security posture by including explicit instructions to avoid default '0.0.0.0' bindings, recommending loopback '127.0.0.1' instead.
- [EXTERNAL_DOWNLOADS]: While the skill references external packages (
flide,torchvision) and models from Hugging Face, these are standard requirements for the stated purpose. The instructions mandate obtaining user authorization before performing any downloads. - [COMMAND_EXECUTION]: The skill executes a local preflight script (
scripts/preflight.py) and provides CLI command examples. The preflight script is safe as it only inspects system metadata (Python version, OS, presence of binaries likenvidia-smi) without performing network or sensitive file operations.
Audit Metadata