Security audit

Frontend Design Extractor

Security checks for vulnerabilities and agentic risk

Overview

This skill is a UI design-system helper with broad but disclosed repository scanning and UI-only editing guidance.

Before installing, understand that this skill may scan a frontend repository and may guide UI-only code edits if you choose a refactor workflow. Use it on repositories where broad UI inspection is acceptable, and review the proposed refactor plan before allowing changes.

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

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
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (4)

Description-Behavior Mismatch

High

Confidence: 97% confidence
Finding: The manifest description limits the skill to extracting reusable UI/UX design systems from frontend codebases and documenting them for reuse, explicitly excluding business logic. However, this section adds a separate 'Refactor from spec' flow with 'actual code changes,' phased execution of edits, and component/page alignment work in a target project, which goes beyond extraction/documentation into code modification.

Intent-Code Divergence

Medium

Confidence: 84% confidence
Finding: The overview states the skill should extract a reusable UI/UX design spec and exclude business logic and domain-specific workflows. Later sections direct the user to request 'actual code changes' and execute phased refactors, which contradicts the earlier extraction-focused framing even if the edits are intended to remain UI-only.

Description-Behavior Mismatch

Medium

Confidence: 93% confidence
Finding: The manifest describes analyzing frontend repositories to extract existing design systems for documentation or migration. In contrast, the greenfield section directs the skill to define foundations, create baseline components, and create page templates 'from blank,' which is generation/scaffolding work rather than extraction from an existing codebase.

Missing User Warnings

Low

Confidence: 84% confidence
Finding: The skill directs scanning an entire repository and creating output folders/files without an explicit consent checkpoint or warning about breadth of access and filesystem modification. In an agent setting, broad repo reads and automatic writes can exceed user expectations, especially in sensitive monorepos or when output paths are not carefully constrained.

Static analysis

No suspicious patterns detected.