understand-image-minimax

• Outcome: Pass
• Latest audit: May 29, 2026 at 8:19 PM UTC
• Version:
  1.0.0

Overview

This skill appears to do what it claims: analyze user-supplied images through the Minimax vision API, with privacy considerations but no hidden or unrelated behavior found.

Install this only if you are comfortable sending image URLs, local image files, base64 image data, and prompts to Minimax for analysis. Avoid using it on private screenshots, documents, personal photos, or confidential work material unless Minimax's data handling terms are acceptable, and use a dedicated Minimax API key where possible.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (3)

Vague Triggers

Medium

• Confidence: 94% confidence
• Finding: The trigger text says the skill must be used whenever a user sends an image or asks about image content, which is overly broad and can cause the agent to invoke an external vision service in ordinary conversations without clear user intent. In this skill’s context, that increases the chance of unnecessary data disclosure because user-provided images or local paths may be sent to a third-party API.

Missing User Warnings

High

• Confidence: 98% confidence
• Finding: The skill documentation does not warn users that submitted images, URLs, or possibly local image content may be transmitted to the Minimax external API for processing. This creates a meaningful privacy and data-handling risk because users may provide sensitive screenshots, personal photos, or internal files without understanding they are leaving the local environment.

Missing User Warnings

Medium

• Confidence: 96% confidence
• Finding: This skill reads a local image file, converts it to base64, and sends the full contents to a third-party Minimax API without any explicit consent prompt, warning, or scoping restriction at the point of transmission. In an agent/skill context, users may provide local paths expecting local analysis, so this creates a real data-disclosure risk if sensitive screenshots, documents rendered as images, or private photos are uploaded off-host unintentionally.

VirusTotal

64/64 vendors flagged this skill as clean.