Ui Ux Pro Max 0.1.0

• Outcome: Review
• Latest audit: May 28, 2026 · 11:19 AM UTC
• Version: 0.1.0

Overview

The skill is a coherent UI/UX helper, but its optional persistence script can write generated files to caller-controlled paths without path validation or confirmation.

Install only if you are comfortable with an optional helper script that can create or overwrite local markdown files. Use the search/design generation in read-only mode by default, avoid --persist unless you intend to write files, and do not pass absolute paths, parent-directory traversal, or untrusted project/page names to the persistence options.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep
• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• MCP Least Privilege: Underdeclared Capability, Wildcard Permission, Missing Permission Declaration
• MCP Tool Poisoning: Hidden Instructions, Unicode Deception, Parameter Description Injection
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands

Findings (7)

Lp3

Medium

• Category: MCP Least Privilege
• Confidence: 89% confidence
• Finding: The skill explicitly instructs the agent to read bundled files and optionally run a local Python script, which implies file-read and potentially file-write/code-execution capabilities despite no declared permissions. This creates a transparency and policy gap: the orchestrator or reviewer may treat the skill as lower risk than it actually is, increasing the chance that local resources are accessed without appropriate gating or user awareness.

Description-Behavior Mismatch

Medium

• Confidence: 95% confidence
• Finding: The skill includes persistence logic that writes generated content to disk under a project-specific folder. For a UI/UX advisory skill, this extends capability from generating guidance to mutating the local filesystem, which increases risk if invoked in an automated agent workflow or against sensitive working directories.

Context-Inappropriate Capability

Medium

• Confidence: 97% confidence
• Finding: The function accepts an arbitrary output_dir and writes files beneath it without validation, allowing the caller to direct writes anywhere the process has permissions. In an agent context, this can be abused to overwrite or plant files in unrelated repositories, config locations, or sensitive paths, which is more dangerous than a normal design-guidance feature warrants.

Vague Triggers

Medium

• Confidence: 81% confidence
• Finding: The activation description is very broad and can match a wide range of normal UI, accessibility, frontend, and design-system requests. Over-broad routing is risky because it can cause this skill to activate in contexts where file access, repo guidance, or local-script suggestions are unnecessary, expanding the attack surface and increasing the chance of unintended capability use.

Vague Triggers

Medium

• Confidence: 85% confidence
• Finding: The README states the skill auto-activates when users request UI/UX work, which is a very broad trigger surface for a powerful code-generating skill. In agent environments, overly loose activation criteria can cause unintended invocation on ordinary frontend discussion, increasing the chance the skill influences tasks the user did not explicitly authorize.

Vague Triggers

Medium

• Confidence: 88% confidence
• Finding: The workflow claims the skill applies to 'any UI/UX task' including build, design, create, implement, review, fix, and improve, which expands the activation scope beyond a safely bounded domain. In practice this can over-match many benign conversations about product ideas or design critique, leading to unsolicited tool behavior or prompt-context takeover by the skill.

Missing User Warnings

Medium

• Confidence: 90% confidence
• Finding: The write path creates directories and writes MASTER/page files immediately once persist=True is passed, with no built-in confirmation, dry-run, or warning in that code path. In agent-driven use, silent persistence can surprise users and lead to unintended modification of project files, especially when combined with arbitrary output directory selection.

VirusTotal

66/66 vendors flagged this skill as clean.

Static analysis

No suspicious patterns detected.