ClawHub
Back to skill

Security audit

Vision Tool

Security checks for vulnerabilities and agentic risk

Overview

This skill appears to be a straightforward local Ollama image-analysis tool with no hidden persistence, credential use, or unrelated behavior.

Install only if you intend to process images with a trusted Ollama instance. Treat submitted images as visible to whatever Ollama service is configured, and avoid sensitive screenshots, documents, or personal photos unless you trust that local service and its model setup.

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
90% confidence
Finding
The skill declares no permissions while its documentation clearly indicates network use via Ollama's chat API. This creates a transparency and policy-enforcement gap: users or the platform may assume the skill is local-only when it actually transmits data to a service, which is especially sensitive for image inputs.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The documentation says the skill reads and analyzes images through the Ollama chat API, but it does not clearly warn users that image contents are transmitted for processing. If users supply sensitive screenshots, personal photos, or documents, they may unknowingly expose private data to a local or remote inference service.

Static analysis

No suspicious patterns detected.