ClawHub
Back to skill

Security audit

vision

Security checks for vulnerabilities and agentic risk

Overview

This skill is a straightforward Google Gemini image-analysis helper, but users should know it sends selected images and questions to Google Vertex AI.

Install this only if you are comfortable sending the images you select, along with your question, to Google Vertex AI using your Google project credentials. Avoid using it on screenshots or documents containing secrets, private records, credentials, or other sensitive content unless that external processing is acceptable.

SkillSpector

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
87% confidence
Finding
The skill documentation describes invoking a local Python script that reads image files or URLs, uses environment variables for cloud credentials, and sends data over the network to Google Vertex AI, yet no permissions are declared. This creates a transparency and policy-enforcement gap: an agent or reviewer may underestimate the skill's ability to access local files, secrets, and external services, increasing the chance of unintended data exfiltration from sensitive images or URLs.

Missing User Warnings

Medium
Confidence
89% confidence
Finding
The script uploads user-supplied images and prompts to an external cloud service, but it provides no explicit runtime disclosure or consent prompt before transmission. In a skill context, users may assume local-only processing and unintentionally send sensitive screenshots, documents, or personal images to a third party.

Static analysis

No suspicious patterns detected.