​
Overview

The Page Capture Agent specializes in detailed analysis of individual web pages by combining visual screenshots with DOM structure analysis.
This agent uses AI vision technology to analyze both the visual appearance and underlying structure of web pages.

​
What Navi Learns

Context

Page purpose, user guidance, and training content

Pages

Page layout, navigation, and interactive elements

​
What Gets Extracted

  • Visual Information: Layout structure, design elements, and interactive components
  • Functional Elements: Form fields, navigation, and action items
  • Training Context: Page purpose, user guidance, and common issues
  • Structured Data: Element selectors, workflow steps, and context relationships

​
Tips for Best Results

Provide extra notes: Add a note to the screenshot to help Navi understand the page.
  • Ensure the page is fully loaded before capture
  • Set the page to its most representative state
  • Clear any overlays or popups that might interfere
  • Focus on key pages in important user workflows