ClawHub

Image Creator

Security checks across malware telemetry and agentic risk

Overview

This is an instruction-only image generation helper with ordinary privacy cautions around uploaded images and broad activation terms.

Before installing, expect this skill to handle many image-related requests and to process uploaded reference images through image analysis and generation tools. Avoid uploading private or sensitive photos unless you have consent from the people shown, and be aware that city, landmark, or brand lookups may use external search when extra details are needed.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Vague Triggers

Medium
Confidence
93% confidence
Finding
The trigger list is very broad and includes generic terms like "figure," "portrait," "city," "weather," and "image generation," which can cause the skill to activate in conversations where the user did not intend to invoke it. In this skill, unintended activation matters because the workflow encourages collecting user images and sending prompts or reference files to external analysis/generation tools, increasing the chance of privacy leakage or confusing, unsafe tool use.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill directs the agent to use `images_understand` on user-provided reference images and `gen_images` with user prompts, but it does not warn users that their images and text may be transmitted to downstream tools or services. This creates a transparency and privacy risk, especially because users may upload personal photos for portraits, chibi figures, or ads without realizing they are being processed externally.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal