ClawHub
Back to skill

Security audit

提示词美学

Security checks across malware telemetry and agentic risk

Overview

This skill is a prompt-template helper for GPT-image2 image prompts and does not show hidden execution, data access, persistence, or destructive behavior.

Install only if you want a Chinese-language GPT-image2 prompt generator. Be aware it may activate on generic prompt-writing phrases, so use explicit wording when you want non-image prompt help.

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 (1)

Vague Triggers

Medium
Confidence
93% confidence
Finding
The trigger list includes broad, everyday phrases such as '写个提示词', '做提示词', and 'prompt me', which can cause the skill to activate outside its intended GPT-image2 image-generation context. This can lead to unintended interception of unrelated user requests, prompt-routing errors, and reduced reliability of the hosting agent.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.