screenshot-ux-auditor

Security checks across malware telemetry and agentic risk

Overview

This is a disclosed local UX-audit helper with a small CSV-generation script and no evidence of hidden access, networking, persistence, or destructive behavior.

Before installing, expect this skill to guide the agent’s UX review and optionally generate a CSV from issue data. Provide only screenshots and product context you are comfortable sharing with the agent, and choose the CSV output path carefully to avoid overwriting an existing file.

SkillSpector

By NVIDIA

Vulnerability Patterns

Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
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)

Tp4

High

Category: MCP Tool Poisoning
Confidence: 95% confidence
Finding: The skill advertises screenshot-based UX auditing, but the observed behavior reportedly only reformats an existing JSON issue list into CSV and does not perform screenshot analysis or audit generation. This mismatch is dangerous because users and downstream agents may trust the skill to analyze screenshots and make decisions based on output that was never actually derived from the provided UI evidence.

Vague Triggers

Medium

Confidence: 87% confidence
Finding: The trigger examples include very broad, everyday phrases such as '审查这个界面' and '页面哪里有问题', which can overlap with normal conversation and cause unintended skill activation. In an agent ecosystem, overly generic triggers can route unrelated user requests into this skill, leading to incorrect handling of screenshots or files and increasing the chance of accidental processing beyond the user's intent.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal