← sources

How do you use a browser alongside your coding agents?

source: reddit (reddit:ClaudeAI)published 2026-10-20 23:11ingested 2026-10-21 02:36

retained text

Curious how people here work with their browser and coding agents like Claude Code or Codex. Do you keep the agent in a terminal next to the browser and check what it did there, or do you let it drive the browser itself?

I'm building a desktop browser that goes the other way: the terminal lives inside the browser, and the agent sits on the side while you browse, with access to the page. Think DIA, but with whatever CLI you already use. Right now I'm working on security and sandboxing, so you can limit what the agent can touch. It's a fork of Orca, and I used Claude to plan the work and write a very good chunk of it's code.

What would you need from a setup like that before you'd trust it with your browser? Which parts of your workflow would it have to cover?

document fetched 2026-10-21 00:20 · canonical https://reddit.com/r/ClaudeAI/comments/1xb750n/how_do_you_use_a_browser_alongside_your_coding