ClawHub
Back to skill

Security audit

装修风格选择与搭配指南

Security checks for vulnerabilities and agentic risk

Overview

This is a Markdown-only home decoration style guide with some practical renovation caveats, but no code, persistence, credential access, or hidden agent-control behavior.

Before installing, understand that this skill is a decorating advice corpus, not a licensed contractor, engineer, lawyer, or financial adviser. For demolition, balcony changes, waterproofing, structural work, code compliance, or property-buying claims, verify locally and consult qualified professionals.

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 (3)

Vague Triggers

Medium
Confidence
91% confidence
Finding
The activation scope is broad enough to trigger on many ordinary home-decoration queries, which can cause the skill to intercept requests beyond a narrowly intended context. Over-broad activation is a security and safety concern because it increases the chance of unintended instruction takeover, misrouting, and suppression of other more appropriate skills or default system behavior.

Vague Triggers

Medium
Confidence
93% confidence
Finding
The phrase instructing users to reply with a very common term (“电视墙”) functions as a broad trigger that could be activated during normal conversation, increasing the chance of unintended skill invocation or routing. In a content/reference file for a home-decoration skill this is not overtly malicious, but it still creates unnecessary ambiguity because the trigger is generic and tightly coupled to ordinary user intent.

Missing User Warnings

Medium
Confidence
90% confidence
Finding
This section recommends removing balcony sliding doors in a home-renovation guidance skill without any visible qualification about building type, façade rules, thermal/energy implications, waterproofing, or the need for property-management/engineering approval. In renovation context, users may treat the advice as actionable; if applied to buildings where the door assembly affects enclosure performance or regulated exterior changes, it can create safety, compliance, moisture, and habitability risks.

Static analysis

No suspicious patterns detected.