Summary of an article published on The Web Scraping Club on 2024-03-05 (free to read). Read the full article
Rotating IPs or using a privacy browser does not hide who you are. Websites read a device fingerprint: operating system, location, plugins and other values. If those values change between sessions or contradict each other, anti-bot systems notice. That matters for scraping sites behind Cloudflare, and for anyone running many accounts that must look like different real users.
I asked Tamas, CEO of Kameleo, to explain what an anti-detect browser does under the hood. We cover how these tools conceal and keep a fingerprint stable across sessions, and why Playwright, built for testing, and add-ons like Puppeteer Stealth struggle against major anti-bot systems. We then go through what a vendor has to maintain: custom-built browsers, constant updates, cloud-reusable profiles, support, and a local API for Selenium, Puppeteer and Playwright. We also look at the planned scraping-only package, where you pay mostly for the browser and skip profile reuse and the GUI.
What’s inside
- Why a consistent fingerprint is more than a mask: the values have to hold up under deeper inspection, which regular tools do not manage.
- What “built-in browser” means: the vendor modifies the native C++ code and ships its own builds, such as Junglefox and Chroma in Kameleo.
- Why maintenance never ends: Chrome updates every 4 weeks and the vendor has to track new fingerprinting research.
- Which scraping jobs need profile reuse, like marketplaces with logins and browser history, and which can do without it.
In the full article: Details of the planned scraping-only package, the platform availability notes, and the full list of vendor responsibilities Tamas describes.