ClawHub
Back to skill

Security audit

UIUX设计助手(免费版)

Security checks for vulnerabilities and agentic risk

Overview

This UI/UX advice skill is not malicious, but it asks for command execution and file-writing abilities that are broader than its stated design-guidance purpose.

Review this skill before installing. It appears to be a UI/UX design-helper Markdown skill with no embedded malware or persistence, but grant it only if you are comfortable with an agent potentially running commands or writing project files. Prefer using it for advice and generated CSS tokens, and require explicit confirmation before any command execution or file changes.

SkillSpector

Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Behavioral ASTexec() Call, eval() Call, Dynamic Import
  • 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
Findings (4)

Description-Behavior Mismatch

Medium
Confidence
94% confidence
Finding
The skill is presented as a UI/UX advisory tool, but the documentation broadens it into operational modes such as create/modify/delete/import/export/save and structured execution flows. This capability expansion can mislead the agent into treating a design-advice skill as an action-taking tool, increasing the risk of unintended file changes or command-driven behavior beyond user expectations.

Description-Behavior Mismatch

Medium
Confidence
91% confidence
Finding
The file states that no extra scripts or external APIs are needed, yet elsewhere advertises EXEC-based operation, browser dependencies, and network troubleshooting. These contradictions weaken operator trust and can hide actual execution/network behavior, which is dangerous because users may approve the skill under false assumptions about what it can do.

Context-Inappropriate Capability

Medium
Confidence
97% confidence
Finding
Declaring exec for a basic UI/UX guidance skill is excessive and violates least privilege. If the agent can execute shell commands in a skill that only needs to generate recommendations and text tokens, that unnecessary capability materially increases the blast radius of prompt misuse, future edits, or agent misinterpretation.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill declares exec and write capabilities without a clear warning that it may run commands or modify project files. In context, this is more dangerous because the skill markets itself as a benign design assistant; users may reasonably expect advice-only behavior and grant use without realizing it can perform side-effecting actions.

Static analysis

No suspicious patterns detected.