ClawHub
Back to skill

Security audit

DESIGN.md UI Designer

Security checks for vulnerabilities and agentic risk

Overview

This skill is a broad but coherent UI and landing-page design workflow with no hidden code execution, credential handling, or persistence beyond normal repo edits.

Install this if you want Codex to take an opinionated role in UI, landing-page, SEO, policy-page, and media polish. Expect it to create or edit repo files, generate or request image assets when useful, and apply quality gates before final delivery. Because the trigger is broad, review proposed changes carefully before publishing or deploying public pages.

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
  • 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 (2)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The trigger description is unusually broad and bundles many generic UI/design terms such as landing page, visual redesign, design tokens, and media assets. This can cause the skill to activate in unrelated contexts, exposing users or downstream agents to a powerful workflow that may create or modify public-facing assets, marketing copy, and deployment-related pages when the user did not specifically request this skill.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
89% confidence
Finding

The default prompt is framed as a broad, always-on instruction set covering multiple domains and actions without clear scope boundaries or invocation constraints. This can cause the agent to over-apply the skill outside intended contexts, increasing the risk of inappropriate autonomous behavior, weak user-intent matching, and accidental use of privileged design or publishing workflows.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.