kernel-microbenchmark
Pass
Audited by Gen Agent Trust Hub on Aug 28, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides benchmarking templates using established libraries like PyTorch, FlashInfer, and vLLM's internal distributed utilities. All external references and packages (e.g., pandas, torch, flashinfer) are standard in the machine learning and high-performance computing ecosystem. The scripts perform local performance measurements and distributed synchronization for benchmarking purposes without any detected malicious patterns such as credential exfiltration, obfuscation, or unauthorized remote code execution.
Audit Metadata