ClawHub

Vx Usage

Security checks across malware telemetry and agentic risk

Overview

This skill is a documentation-only guide for using vx, with clear auto-install behavior and no executable payload, though users should be aware it can steer agents toward environment-changing vx commands.

Install only if you want agents to use vx conventions. Before allowing commands like vx install, vx sync, vx setup, vx lock, vx ai setup, Git/GitHub mutations, or CI/MCP configuration changes, confirm the project is vx-managed and that you accept local installs, cache changes, project-file updates, or token-bearing integrations.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • 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
Findings (2)

Vague Triggers

Medium
Confidence
91% confidence
Finding
The skill’s activation description is broad enough to trigger on common topics like Git/GitHub operations or cross-platform setup, even when the user did not ask to use vx. In an agent setting, that can inappropriately steer behavior toward this skill’s guidance and tool choices, increasing the chance of irrelevant command execution patterns or unintended environment changes.

Missing User Warnings

Medium
Confidence
88% confidence
Finding
The skill repeatedly instructs agents to run `vx` commands that auto-install tools and later recommends setup commands like `vx setup`, `vx sync`, and `vx ai setup`, but it does not prominently warn that these actions can download software, modify caches, create project files, or change local environments. In autonomous or semi-autonomous agent use, that omission can lead to silent system or repository modifications beyond the user’s explicit intent.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal