ClawHub
Back to skill

Security audit

AI Product Visual Prompt Builder

Security checks across malware telemetry and agentic risk

Overview

This is a text-only prompt-writing skill for product visuals and does not request file access, credentials, commands, persistence, or high-impact permissions.

Install this as a prompt-writing helper, not as a tool that operates image generators directly. If you use the optional external testing site, review that service separately and avoid uploading confidential product images, unreleased campaign materials, or sensitive brand assets unless you trust its terms and data handling.

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
92% confidence
Finding
The description is broad enough to match many generic creative-help requests, which can cause the agent to invoke this skill outside its intended niche. Over-broad routing increases the chance of prompt hijacking, poor task selection, or accidental suppression of more appropriate skills, though the skill itself is text-only and does not request secrets or code execution.

VirusTotal

62/62 vendors flagged this skill as clean.

View on VirusTotal