ClawHub
Back to skill

Security audit

专业物体识别技能。输入图片，返回场景、物体名称、类别、品牌、材质、价格和置信度

Security checks across malware telemetry and agentic risk

Overview

ObjectVision is an instruction-only image recognition skill with expected privacy considerations but no executable code, persistence, credential access, or hidden behavior.

Install only if you are comfortable sending image URLs for vision analysis. Avoid private photos, internal URLs, workplace screenshots, or images showing sensitive possessions or locations unless that exposure is acceptable, and treat brand and price estimates as uncertain model inferences.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (2)

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The skill accepts arbitrary image URLs and implies the system will fetch and analyze remote images, but it does not warn users that submitted URLs and associated image data may be transmitted to external services or cause server-side retrieval. This creates a real privacy and data-handling risk, especially if users submit sensitive or internal URLs without understanding the consequences.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The 'professional' mode explicitly advertises detection of small objects and drawer interiors, which increases the likelihood of extracting sensitive personal, financial, or confidential information from images. Without a privacy warning or usage safeguards, the feature encourages deeper inspection of potentially private spaces and makes the skill more dangerous in context than ordinary object detection.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal