external-cannbot-model-model-infer-multi-stream

Pass

Audited by Gen Agent Trust Hub on Jul 7, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides purely instructional content and technical documentation for optimizing PyTorch models on Ascend NPUs. No malicious commands or suspicious logic were detected.
  • [SAFE]: All referenced APIs (e.g., torch.npu.Stream, npu_stream_switch, record_event) are standard components of the Huawei CANN (Compute Architecture for Neural Networks) and PyTorch NPU stacks.
  • [SAFE]: No obfuscation, hidden URLs, or data exfiltration patterns were found. The skill uses standard environment variables for configuration and does not access sensitive system files.
  • [SAFE]: The skill documentation and examples are consistent with its stated purpose of improving inference performance through stream overlap and core management.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 7, 2026, 02:39 AM
Security Audit — agent-trust-hub — external-cannbot-model-model-infer-multi-stream