Bots in browsers
The painted mark for HeadlessChrome
Placeholder: mark still to be painted

HeadlessChrome

Google Chrome running without a window, driven by a program that tells it which pages to open and what to do there. It is used for automated testing, screenshots, making PDFs, monitoring and link previews, and for scraping. Nobody in particular runs it.

Worth knowing: Out of the box it calls itself “HeadlessChrome”, which is why scrapers that want to pass as people usually change its name (Castle). Headless mode first came with Chrome 59, in 2017 (Chrome).

How it makes money: Driving a browser is sold by the page: the scraping company Zyte charges from $1.01 per 1,000 pages rendered in a browser up to $16.08 for the hardest sites, against $0.13 per 1,000 plain pages. (zyte.com)

On this site: Almost always the homepage and never the rules, from rented servers: tools taking screenshots of the site, checking it is up or testing it, rather than reading it.

Why? Many sites turn away programs that admit to being programs, so dressing up as a browser gets through. Renting servers in the cloud is cheap and each one comes with a fresh address; borrowing someone’s home connection costs more, but is the hardest to tell from a person. In August 2025 Cloudflare reported that an AI search company, once its named crawlers were blocked, carried on with an unnamed one that said it was Chrome on a Mac and kept changing its addresses (report). What they sell →