​
What can you do with it?

The /image-understanding command enables you to analyze images and extract detailed information using AI vision capabilities. You can identify objects and people, describe scenes and activities, analyze visual elements like colors and composition, extract text from images, and answer specific questions about image content.

​
How to use it?

​
Basic Command Structure

/image-understanding [image-file] [query]

​
Parameters

Required:
  • image-file - The image to analyze (URL or uploaded file)
Optional:
  • query - Specific question or analysis request about the image
  • format - Output format: descriptive text or json (defaults to descriptive text)
  • detail-level - Analysis detail: basic, detailed, or comprehensive

​
Response Format

The command returns: 
{
  "output": "detailed description or analysis result",
  "objects_detected": ["list of identified objects"],
  "text_found": "any text extracted from the image",
  "metadata": {
    "dominant_colors": ["color1", "color2"],
    "image_type": "photograph/illustration/diagram"
  }
}

​
Examples

​
Basic Usage

/image-understanding
image-file: vacation-photo.jpg
Provides a general description of what’s shown in the image.

​
Advanced Usage

/image-understanding
image-file: product-display.png
query: Describe all products visible and their arrangement
format: json
detail-level: comprehensive
Analyzes a product display image with detailed information about each item and layout in JSON format.

​
Specific Use Case

/image-understanding
image-file: chart.png
query: What trends are shown in this chart?
Analyzes a chart image to identify and explain the data trends displayed.

​
Notes

Supports object and scene recognition, text extraction (OCR), color and composition analysis, facial detection (no identification), activity and action recognition, and diagram and chart interpretation.