Skip to main content
The Vision skill lets the assistant analyze images, screenshots, and visible page content. It can describe what it sees, extract text, and answer questions about visual content.

​
Use Cases

  • Analyzing screenshots and images for quick understanding without manual inspection.
  • Transcribing visual content from a page into text you can reference or store.

​
Call it from the slash command system

Type /vision in the composer, select the skill, then attach an image or describe what you want analyzed.

​
Ask the assistant directly

Pro users can also ask the assistant to use the Vision skill in plain language without opening the slash menu.