ClawHub
Back to skill

Security audit

Gemini Image Proxy

Security checks across malware telemetry and agentic risk

Overview

This skill is a straightforward image generator/editor that sends prompts and optional source images to a user-configured API and saves the returned image locally.

Install only if you trust the OpenAI package source and the GOOGLE_PROXY_BASE_URL endpoint. Use a limited API key where possible, and avoid confidential prompts, private images, or regulated data unless the endpoint's logging, retention, and data-handling practices are acceptable.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • 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)

Lp3

Medium
Category
MCP Least Privilege
Confidence
89% confidence
Finding
The skill declares required environment variables under metadata but does not explicitly declare corresponding permissions, creating a mismatch between documented capabilities and the permission model. This can mislead users or host systems about what sensitive resources the skill needs, reducing transparency around access to API secrets.

Missing User Warnings

Medium
Confidence
96% confidence
Finding
The skill encourages users to generate and edit images via a remote OpenAI-compatible endpoint but does not warn that prompts and uploaded source images are transmitted to an external service. Users may unknowingly send sensitive images, proprietary material, or confidential prompt content off-box, creating privacy and data-handling risk.

VirusTotal

55/55 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.