ClawHub
Back to skill

Security audit

huawei-cloud-ascendc-operator-performance-optim

Security checks across malware telemetry and agentic risk

Overview

The skill is mostly a coherent AscendC profiling and optimization aid, but it includes overbroad activation language and unsafe permission guidance that should be reviewed before use.

Install only for AscendC/CANN operator work. Avoid following the chmod 777 guidance; use a private profiling output directory with restrictive permissions instead. Run the included scripts only in an operator workspace where deleting OPPROF_* profiling outputs is acceptable.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Vague Triggers

Medium
Confidence
88% confidence
Finding
The trigger list includes generic terms such as "performance", "custom operator", and "operator optimization" that can match many unrelated requests, causing the agent to invoke this skill outside its intended AscendC/Huawei NPU context. This is primarily a scope-control issue rather than direct code execution risk, but it can lead to inappropriate tool use, misleading guidance, or unnecessary exposure of local files/performance data during profiling workflows.

Missing User Warnings

Medium
Confidence
97% confidence
Finding
The troubleshooting guidance recommends `chmod 777 /tmp/opprof`, which makes the profiling directory world-writable without any warning or scope limitation. On multi-user systems this can enable unauthorized modification, file planting, symlink attacks, or tampering with profiling outputs, and the skill context does not require such broad permissions to function.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.