Why raw DOM is a bad control surface
Raw DOM exposes a huge amount of irrelevant state. That increases prompt size and makes action planning brittle, especially on pages with marketing overlays or personalization modules.
Agent browsingSEO landing
Prepare browser agent content for model use
Browser agents do not need every node in the DOM. They need enough structured page context to decide where to click, what to summarize and when to call tools.
Primary use
Give browser agents compact page context instead of forcing them to reason over raw DOM noise.
Recommended flow
Fetch, clean, measure tokens, then hand consistent Markdown to agents or retrieval systems.
Next step
Use the Playground to compare raw HTML against optimized output before integrating the API.
Better browser-agent context
The better pattern is to keep browser automation for navigation and interaction, then hand the model a cleaned content view for reasoning. That separates action state from reading state.
Internal links
Related technical paths
Agent access
Reader API for AI systems
Expose web content to agents and retrieval systems through a reader-style API that prioritizes clarity over browser markup.
Cost control
Reduce HTML token usage before RAG or agents
Shrink token spend by converting bloated HTML into compact Markdown before chunking, prompting or embedding.
Core positioning
AI ingestion infrastructure for websites
Normalize public web content into consistent, low-noise context that AI systems can index, retrieve and reason over.