mindspeed-mm-pipeline

Pass

Audited by Gen Agent Trust Hub on May 19, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides instructions for identifying and terminating stale training processes using standard shell commands (ss, ps, grep, kill). These are documented for local environment maintenance and process cleanup.
  • [EXTERNAL_DOWNLOADS]: References official repository and documentation links to the vendor's GitCode page (gitcode.com/ascend/MindSpeed-MM). These are legitimate resources for the MindSpeed-MM framework.
  • [SAFE]: Documents required Docker runtime configuration, specifically the use of --privileged for NPU hardware access and --ipc=host for shared memory allocation. These parameters are standard requirements for deep learning workloads on Ascend NPU devices.
Audit Metadata
Risk Level
SAFE
Analyzed
May 19, 2026, 06:27 AM
Security Audit — agent-trust-hub — mindspeed-mm-pipeline