ClawHub
Back to skill

Security audit

GPT Image 2 产品图生成

Security checks across malware telemetry and agentic risk

Overview

This skill coherently generates product images through an external AI service, with expected local API-key setup and user-directed image upload/download behavior.

Install only if you are comfortable sending selected product reference images and prompts to AI Hive, storing an AI Hive API key locally, and saving generated files to the default downloads folder. Use approved source materials for claims such as certifications, dimensions, ingredients, or regulated product details.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (2)

Lp3

Medium
Category
MCP Least Privilege
Confidence
93% confidence
Finding
The skill advertises only a product-image workflow, yet the documented execution path installs dependencies and invokes a Python script with shell, file, environment, and network access while declaring no permissions. This creates a transparency and least-privilege problem: users and platforms cannot accurately assess or constrain what the skill can do, and a compromised or overly broad script could access local data, exfiltrate secrets, or modify files.

Tp4

High
Category
MCP Tool Poisoning
Confidence
97% confidence
Finding
The finding indicates the underlying tooling is a general-purpose AI Hive/OpenAPI client with capabilities far beyond product-image generation, including chat, video generation, model enumeration, user/wallet queries, media upload, browser-based API-key setup, and downloading arbitrary task results. That mismatch is dangerous because it defeats user expectations and policy scoping: a skill invoked for benign product images may silently exercise unrelated privileged features or expose sensitive account and local environment data.

VirusTotal

52/52 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.