ClawHub
Back to skill

Security audit

Hardware Llm Optimizer

Security checks across malware telemetry and agentic risk

Overview

This skill transparently checks local hardware to recommend runnable LLM models, with no evidence of hidden network access, credential access, persistence, or destructive behavior.

Install this only if you want local hardware-based LLM recommendations. Before running it, understand that it will print system details such as CPU, RAM, GPU, VRAM, driver, and OS/WSL information into the agent session; ask for confirmation before hardware probing if your request is only for general advice, and request translation if the Chinese output is not suitable.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (2)

Lp3

Medium
Category
MCP Least Privilege
Confidence
90% confidence
Finding
The skill advertises executable behavior (`python3 .../detect.py`) and hardware auto-detection that necessarily implies local file/system inspection and likely shell access, but the manifest declares no permissions. This creates a transparency and consent gap: a user or platform may invoke a skill that reads local system state or runs commands without an explicit permission contract.

Vague Triggers

Medium
Confidence
83% confidence
Finding
The invocation phrases are very broad (`检测电脑配置`, `大模型推荐`, `LLM优化`) and can cause the skill to activate in situations where the user may only want general advice, not local hardware probing. In a skill that can inspect host hardware and run detection commands, ambiguous activation increases the risk of unintended data access or command execution.

VirusTotal

67/67 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.