Skip to content
Agentic WorkFlow

Text Extraction Node Comparison Guide

What this guide does: Helps you choose the right text extraction node for your specific workflow by comparing the different options available.

Quick Comparison

Section titled “Quick Comparison”
NodeWhen to UseUser Action RequiredBest For
Get Selected TextYou want specific textYes - highlight textFocused analysis, quotes
Get All TextYou want everythingNoFull page analysis, monitoring
Get HTML of Selected TextYou want formatted textYes - highlight textPreserving formatting
Get All HTMLYou want complete structureNoWeb extraction, archiving

Real-World Examples

Section titled “Real-World Examples”

Research Assistant Use Get Selected Text when you want to analyze specific quotes or findings that catch your attention while reading.

Content Monitoring Use Get All Text when you want to automatically check if a webpage has changed since your last visit.

Content Archiving Use Get All HTML when you want to save complete webpages with all their formatting and structure intact.

Decision Helper

Section titled “Decision Helper”

Ask yourself these questions:

  1. Do you want the user to choose what text to extract?

    • Yes → Use Get Selected Text or Get HTML of Selected Text
    • No → Use Get All Text or Get All HTML

  2. Do you need to keep formatting (bold, links, etc.)?

    • Yes → Use Get HTML of Selected Text or Get All HTML
    • No → Use Get Selected Text or Get All Text

  3. Are you processing with AI?

    • Yes → Use text extraction (cleaner for AI)
    • No → Either option works

Performance Guide

Section titled “Performance Guide”

Speed (fastest to slowest):

  1. Get Selected Text ⚡ (only selected content)
  2. Get HTML of Selected Text ⚡ (only selected content)
  3. Get All Text 🐌 (entire page)
  4. Get All HTML 🐌🐌 (entire page with formatting)

Memory Usage (lightest to heaviest):

  1. Get Selected Text 📱 (small selections)
  2. Get HTML of Selected Text 📱 (small selections)
  3. Get All Text 💻 (text only)
  4. Get All HTML 💻💻 (complete page data)
🔍 Technical Details

Browser Compatibility:

  • All nodes work in Chrome, Firefox, and Edge
  • Safari has limited support for selection-based nodes
  • HTML extraction may be slower on older browsers

Common Issues:

  • No text extracted? Page might still be loading - add a delay
  • Missing content? Some text might be loaded by JavaScript after the page appears
  • Too slow? Set length limits or exclude unnecessary page sections

Performance Tips:

  • Use selection-based nodes for better performance
  • Set maximum length limits for large pages
  • Exclude navigation and ads when extracting full pages

Common Workflow Patterns

Section titled “Common Workflow Patterns”

Pattern 1: User Research

  • Step 1: User highlights interesting text
  • Step 2: Get Selected Text
  • Step 3: AI Analysis Node
  • Step 4: Save insights

Pattern 2: Content Monitoring

  • Step 1: Navigate to page
  • Step 2: Get All Text
  • Step 3: Compare with previous version
  • Step 4: Alert if changed

Pattern 3: Content Enhancement

  • Step 1: User selects text of interest
  • Step 2: Get HTML of Selected Text
  • Step 3: AI Improvement Node
  • Step 4: Replace original content

What’s Next?

Section titled “What’s Next?”