ClawHub
Back to skill

Security audit

Ux Design Tool Free

Security checks across malware telemetry and agentic risk

Overview

This UX guidance skill requests command and file-write permissions that its own documentation says it does not need, so it should be reviewed before installation.

Before installing, ask the publisher to remove exec/write permissions or clearly document exactly when they are used. Avoid providing callback_url values or allowing network diagnostic commands unless you explicitly want that behavior in the current workspace.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Intent-Code Divergence

Medium
Confidence
94% confidence
Finding
The skill declares powerful exec/write tools in the manifest while later claiming it is a pure Markdown skill that does not require command execution. That mismatch can cause users or orchestrators to grant broader privileges than the documented behavior suggests, increasing the chance of unintended command execution or file modification if the skill is invoked or later extended.

Intent-Code Divergence

Low
Confidence
87% confidence
Finding
The error-handling section tells the agent/user to perform network troubleshooting actions such as ping/firewall/proxy checks, which is inconsistent with a Markdown-only UX guidance skill. While not directly malicious, this broadens expected operational behavior and may lead an agent with exec access to run unnecessary network commands in environments where such actions are sensitive.

Missing User Warnings

Medium
Confidence
90% confidence
Finding
The skill documents a callback_url parameter and declares exec/write capabilities without warning users about outbound data transfer, file changes, or command execution. In an agent setting, undisclosed network callbacks or local modifications can expose sensitive input data or alter the workspace unexpectedly, especially because the skill’s stated function is only UX guidance.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.