ClawHub
Back to skill

Security audit

X-VPN MCP Skill

Security checks across malware telemetry and agentic risk

Overview

This is a disclosed VPN helper skill whose main risks are expected for its purpose and should be reviewed before use.

Install only if you trust X-VPN and are comfortable with a local VPN daemon and MCP configuration changes. Prefer inspecting the installer and verifying its source before running it, and remember that while connected, traffic may route through the selected VPN region until disconnected or restored.

SkillSpector

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

Missing User Warnings

Medium
Confidence
96% confidence
Finding
The skill instructs the agent to suggest executing a remote installer directly via `sh <(curl ...)`, which pipes downloaded code into a shell without verification, integrity checks, or a clear warning about the risks. In this skill's context, that is more dangerous because the installation is presented as the normal recovery path when tools are unavailable, creating a strong chance that users will run untrusted code from the network with minimal scrutiny.

Missing User Warnings

Medium
Confidence
89% confidence
Finding
The document repeatedly instructs an agent to establish VPN connections and perform web requests through alternate geographic egress without requiring an explicit user-facing notice or confirmation about privacy, routing, and jurisdictional implications. In this skill context, that is a real safety issue because the skill is specifically designed to alter network origin as part of completing tasks, which can expose user activity to third-party VPN infrastructure and create legal, compliance, or expectation-of-consent problems if done silently.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.