ClawHub
Back to skill

Security audit

computer-control

Security checks across malware telemetry and agentic risk

Overview

This skill clearly describes opt-in desktop automation that can view the screen and control mouse/keyboard, with appropriate safety warnings and no bundled executable code.

Install only if you intentionally want an agent to control a desktop session. Run it in a VM, container, or otherwise isolated display, close private windows first, avoid banking or sensitive accounts, and require confirmation for actions with real-world consequences.

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 (1)

Vague Triggers

Medium
Confidence
89% confidence
Finding
The trigger phrase is broad enough to match many ordinary requests about GUI scripting or browser recording, which could cause this high-risk skill to be invoked more often than intended. Because the skill can capture screenshots and generate mouse/keyboard input with cross-process side effects, overbroad invocation increases the chance of unintended desktop control, privacy exposure, or unsafe automation in the wrong context.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.