nnunet-segmentation
Pass
Audited by Gen Agent Trust Hub on May 6, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No security vulnerabilities or malicious patterns were identified in the skill. The instructions and code samples are consistent with legitimate medical image analysis workflows.\n- [COMMAND_EXECUTION]: The skill uses established command-line tools associated with the nnU-Net framework (e.g.,
nnUNetv2_train,nnUNetv2_predict) to perform training and inference, which is the intended functionality.\n- [EXTERNAL_DOWNLOADS]: The skill references installation of official packages (nnunetv2,torch) via standard package managers and points to Zenodo, a well-known and trusted scientific repository, for downloading pretrained models.
Audit Metadata