
From the team
Blog
What we ship, how it performs,
and practical ways to put rtrvr to work.


Playbooks
Choosing a browser-agent stack
A comparison of rtrvr, Browser Use, Skyvern, and Firecrawl across execution, extraction, infrastructure, and finished outcomes.
Read article
Product
How rtrvr handles data, sessions, and browser access
The boundaries around page data, authenticated sessions, cloud runs, retention, and user control.
Read article
Research
Why rtrvr reads the DOM instead of screenshots
How text-first page state changes browser-agent speed, context use, action precision, and failure modes.
Read article
Research
Why context quality decides scraping results
A comparison of raw page volume and structured DOM context for extraction work that models can actually reason over.
Read article
Product
rtrvr v24: export Smart DOM Trees as Markdown
Get a compact, readable representation of a page for downstream models and structured extraction.
Read article
Product
The rtrvr API: browser work from your product
Send a browser task, keep the sources and structured output, and connect the result to the workflow that needs it.
Read article
Research
WhatsApp's US Breakthrough & The Future of Enterprise AI Agents
With WhatsApp's 39% surge among US iPhone users, enterprise automation is moving to mobile. Here's why rtrvr.ai built for this shift months ago.
Read article
Playbooks
Use your browser as a remote MCP server
Expose browser actions to an MCP client while keeping the session and page state in the browser you control.
Read article
Product
Introducing rtrvr Cloud
Run browser tasks away from your laptop, observe the session, and connect completed work to your product.
Read articlePUT IT TO WORK