While building https://t.co/kE8vin6CZt, I keep been thinking:
Search gives agents results.
Browsers give agents interaction.
Scrapers give agents data.
But what agents really need is clean web context they can reason over.
What should the default web input for an AI agent be?
I built Reader because AI agents do not need raw HTML.
They need clean web context.
Reader turns public URLs into Markdown, structured JSON, crawl results, and browser sessions so agents can read, research, extract, and act on web content.
https://t.co/oajOsr3hda