ascend-dmi
Warn
Audited by Snyk on Apr 18, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).
- Potentially malicious external URL detected (high risk: 0.90). The skill's environment-setup includes an explicit runtime installation command that downloads and executes remote code (wget https://ascend-repo.obs.cn-east-2.myhuaweicloud.com/MindX/MindX%207.3.0/Ascend-mindx-toolbox_7.3.0_linux-aarch64.run and then runs the installer), which the skill may invoke to satisfy a required dependency (MindCluster ToolBox/ascend-dmi), so this URL poses a high-risk external execution dependency.
MEDIUM W013: Attempt to modify system services in skill instructions.
- Attempt to modify system services in skill instructions detected (high risk: 0.90). Yes — the skill explicitly instructs the agent to run hardware-management and high-risk commands on the host (ascend-dmi for resets, stress tests, device checks, fuser on /dev/davinci*, etc.), which modify hardware and system state, can disrupt services and likely require elevated privileges.
Issues (2)
W012
MEDIUMUnverifiable external dependency detected (runtime URL that controls agent).
W013
MEDIUMAttempt to modify system services in skill instructions.
Audit Metadata