视觉笔记卡片生成器

• Outcome: Pass
• Latest audit: May 29, 2026, 8:26 PM
• Version:
  1.0.0

Overview

This skill is a coherent visual note-card generator, but its generated HTML can contact third-party services for fonts and browser export.

Install if you are comfortable with generated HTML loading Google Fonts and a CDN export script. Avoid using confidential source material unless you remove or vendor those remote dependencies first, and confirm the desired output language when prompting.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• MCP Tool Poisoning: Hidden Instructions, Unicode Deception, Parameter Description Injection
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (3)

Description-Behavior Mismatch

Medium

• Confidence: 96% confidence
• Finding: The template imports Google Fonts at render time, which means opening the generated HTML causes outbound network requests and leaks viewer metadata such as IP address, user agent, and referrer context to a third party. This contradicts the skill's description of producing a self-contained artifact and creates both privacy risk and supply-chain dependence if the remote asset changes or becomes unavailable.

Description-Behavior Mismatch

Medium

• Confidence: 99% confidence
• Finding: Loading html2canvas from a third-party CDN introduces a script supply-chain risk: any compromise of the CDN, upstream package, or delivery path would execute arbitrary JavaScript in the page context. Because this page handles export and DOM access, a malicious script could exfiltrate content, alter output, or perform unexpected network activity whenever the generated file is opened.

Missing User Warnings

Medium

• Confidence: 88% confidence
• Finding: The README says output is self-contained while also noting dependence on Google Fonts and html2canvas CDN, but it does not clearly warn users that opening generated HTML may trigger outbound network requests. This can leak metadata such as IP address, user agent, and usage timing, and it weakens assumptions about offline or privacy-preserving use.

VirusTotal

66/66 vendors flagged this skill as clean.