Back to skill

Security audit

装修避坑问答

Security checks across malware telemetry and agentic risk

Overview

This is a renovation advice skill made of markdown guidance files, with no code, credential access, persistence, or automatic actions.

Before installing, understand that this skill gives direct renovation and home-buying advice but does not verify local building codes, contractor terms, or product safety details. Use extra caution for electrical, waterproofing, structural, chemical finishing, and major purchase decisions, and treat the broad triggers as a usability issue rather than hidden behavior.

SkillSpector

By NVIDIA
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
Confidence
87% confidence
Finding
The trigger list includes highly generic phrases such as “要注意什么”, “有什么坑”, and “哪个好”, which can match many unrelated user queries outside the home-renovation context. This can cause unintended activation and prompt hijacking-by-context, where the skill answers in a domain it is not meant for, increasing the chance of misleading or unsafe guidance.

Missing User Warnings

Medium
Confidence
93% confidence
Finding
The advice to search for unfinished wood furniture and 'sand and paint it yourself' encourages a DIY finishing task that can expose users to paint/finish fumes, dust, and skin contact hazards without any safety guidance. In a home-renovation skill aimed at beginners, omission of ventilation, PPE, and product-safety warnings materially increases the chance of unsafe use.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.