ClawHub
Back to skill

Security audit

ui-design-system

Security checks for vulnerabilities and agentic risk

Overview

This is a Markdown-only UI design guidance skill with minor usability and network-placeholder caveats, not evidence of malicious behavior.

Reasonable to install for UI design assistance. Be aware that broad triggers may activate it more often than intended, and avoid or replace the external placeholder-image URL when working in privacy-sensitive or offline environments.

A.I.G

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

SkillSpector

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 (4)

Tp2

High
Category
MCP Tool Poisoning
Confidence
85% confidence
Finding

Mixing characters from multiple Unicode scripts in a single identifier is a common technique to create visually ambiguous tool names.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The trigger list includes broad terms like 'theme', 'aesthetic', 'UI风格', and '设计系统', which can match many ordinary conversations and cause unintended skill invocation. Over-broad activation expands the skill’s operational surface area, making it easier for unrelated prompts to pull in this skill’s instructions and potentially override more appropriate context or create prompt-routing confusion.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
80% confidence
Finding

The document requires text casing to be "Exclusively UPPERCASE" for both structural headers and metadata. In some languages and locales, forced uppercase can reduce readability or behave poorly without user choice, and the file does not mention any opt-in or localization exception.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Low
Category
Not specified by scanner
Confidence
91% confidence
Finding

The skill instructs consumers to fetch placeholder images from an external domain (picsum.photos) when assets are unavailable. This creates network-dependent behavior not disclosed in the skill metadata and can leak usage metadata (IP, timing, referer) or violate environments that require offline/local-only assets. In a design-system skill this is not catastrophic, but it is still unsafe because it normalizes implicit third-party resource loading.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.