ClawHub
Back to skill

Security audit

Design MD Generator

Security checks across malware telemetry and agentic risk

Overview

This skill transparently opens a user-provided website to extract visual design information and create local design documentation, with privacy and browser-sandbox cautions.

Install only if you want an agent to inspect websites and generate local design docs from them. Use public or authorized URLs, avoid authenticated or sensitive pages unless intentional, review generated DESIGN.md files before reusing them as agent context, and prefer an isolated environment when analyzing untrusted sites.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • 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
89% confidence
Finding
The README explicitly states that the skill captures screenshots and DOM snapshots from arbitrary live websites, but it provides no warning about privacy, sensitive data exposure, authentication state, or how captured content is stored and handled. In an agent context, this is risky because users may point the tool at internal apps, logged-in pages, or sites containing personal or confidential information, causing unintended collection of sensitive visual and DOM data.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The README explicitly describes loading arbitrary URLs, taking screenshots, collecting DOM snapshots, and extracting CSS tokens, but it provides no guardrails about authorization, privacy, or handling authenticated/internal pages. In an agent context, this can normalize inspection of sensitive sites and cause accidental capture of confidential content, session-bound UI, or private design artifacts from targets the user should not scrape.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal