ClawHub
Back to skill

Security audit

Picture it!

Security checks across malware telemetry and agentic risk

Overview

This skill is a coherent image generation and editing helper that clearly discloses its fal.ai uploads, API key use, and possible costs.

Install only if you are comfortable using a global npm CLI with fal.ai. Use a limited FAL key, do not paste secrets into chat, and remember that generate, edit, remove-bg, and upscale can upload images/prompts and incur charges; prefer local-only commands when privacy or cost matters.

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
96% confidence
Finding
The trigger text is extremely broad and is designed to activate on almost any image-related request, including routine tasks like cropping, resizing, adding text, or general visual content creation. Overbroad activation increases the chance this skill is invoked unnecessarily, causing unintended data transmission to fal.ai, unnecessary cost-incurring model calls, and expansion of the tool's effective permission surface.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.