ai-for-science-ankh-ascend-npu-skill

Pass

Audited by Gen Agent Trust Hub on May 19, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is a collection of migration guides and verification scripts for protein language models. It uses standard PyTorch and HuggingFace Transformers patterns adapted for Huawei Ascend NPU (torch_npu).
  • [COMMAND_EXECUTION]: Provides standard shell instructions for setting up the Huawei CANN environment and executing model verification scripts. No suspicious or arbitrary command execution patterns were found.
  • [DATA_EXPOSURE]: The scripts interact with local file systems only to load model weights and configurations. There are no hardcoded credentials or external data exfiltration patterns.
  • [REMOTE_CODE_EXECUTION]: No remote script downloads or dynamic code execution via eval/exec on untrusted data were detected. All dependencies are standard libraries for the target environment.
Audit Metadata
Risk Level
SAFE
Analyzed
May 19, 2026, 06:27 AM
Security Audit — agent-trust-hub — ai-for-science-ankh-ascend-npu-skill