nemo-mbridge-perf-moe-comm-overlap
Pass
Audited by Gen Agent Trust Hub on Jul 10, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides standard technical documentation, configuration parameters, and performance results for deep learning training optimizations.
- [COMMAND_EXECUTION]: Includes a bash command example for running a performance benchmark script ('scripts/performance/run_script.py') using the 'uv' package manager. This is a standard development workflow and does not involve remote code execution or untrusted inputs.
Audit Metadata