BrowsingBee

BrowsingBee bot

When a user asks Claude or ChatGPT to open a page through BrowsingBee, our hosted browser fetches that one page on their behalf. This page explains how it behaves and how to opt out.

How it identifies itself

Every request carries this user agent:

Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/129.0.0.0 Safari/537.36 BrowsingBee/1.0 (+https://browsingbee.com/bot)

Requests come from Cloudflare’s network (the browser runs on Cloudflare Browser Rendering).

What it does and does not do

  • Fetches single pages on explicit user request. It does not crawl, follow links on its own, or revisit pages on a schedule.
  • Reads robots.txt once an hour per site and respects Disallow rules for BrowsingBee and for * on page reads, link lists and data extraction.
  • Never logs in, submits forms, clicks, or enters any data into a page.
  • Does not store page content. Screenshots a user asks for are deleted after 24 hours.
  • Is rate limited per user (20 page loads a minute, 300 or 5,000 a month).

Opt out

Add this to your robots.txt:

User-agent: BrowsingBee
Disallow: /

The change takes effect within an hour. To block a section only, list that path instead of /. If you would rather we never fetch your domain at all, including screenshots, email hey@browsingbee.com with the domain and we add it to our deny list.

Abuse

If you see traffic from BrowsingBee that does not match the above, send the user agent, time and path to hey@browsingbee.com. We can trace every request to an account and suspend it.