ClawHub
Back to skill

Security audit

GPT Image 2 海报生成

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed AI Hive poster/image-generation workflow with expected API-key, upload, network, and output-file behavior, though its helper script contains broader unused client code.

Install this only if you are comfortable giving the helper an AI Hive API key, sending prompts and chosen reference files to AI Hive, and saving generated outputs locally. Treat the bundled helper as broader than the poster examples internally, but the active skill command path is focused on image generation.

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
91% confidence
Finding
The skill declares no permissions, yet its documented execution path uses shell commands, network access, environment variables, and file operations via `pip3 install` and `python3 ... imagegen.py`. This creates a transparency and least-privilege problem: users and reviewers cannot accurately assess what the skill can do, and a generic helper script with these capabilities could be repurposed for unintended actions.

Tp4

High
Category
MCP Tool Poisoning
Confidence
96% confidence
Finding
The skill is presented as a narrowly scoped poster-generation tool, but the underlying behavior reportedly supports broader AI Hive operations such as model enumeration, media upload, user/wallet queries, generic task handling, and even non-poster generation modes. That mismatch is dangerous because it can mislead users into granting trust to a specialized design skill that actually exposes a much broader operational surface, increasing the chance of data exfiltration, misuse of account resources, or execution outside the user's intent.

VirusTotal

61/61 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.