Summary of an article published on The Web Scraping Club on 2026-06-14 (free to read). Read the full article
Linux is where most scraping infrastructure lives: cloud servers, Kubernetes, CI pipelines, VPSs. Kameleo, an anti-detect browser, only ran on Windows and macOS for years. Teams that wanted its fingerprints in production ended up with Windows VMs next to their Linux stacks on AWS, or X11-over-SSH tunnels. Both are hard to maintain and hard to scale.
We look at Kameleo Docker, the containerized version of Kameleo. I explain why it exists, with comments from founder and lead engineer Barnabas Szenasi, whom I met at Prague Crawl 2026. Then I go through the architecture: the Local API, the Chroma and Junglefox kernels, profiles and Docker volumes. I cover the main features, including fingerprint masking, proxy and geo consistency, multi-profile isolation and the extras of the Linux image. Finally, I set it up and connect Playwright in Python to a JavaScript-rendered sandbox page from the Scraping Course.
What’s inside
- How the container splits the work: browsers and fingerprinting inside, your automation scripts outside, connected over a CDP WebSocket.
- Why Kameleo does not spoof TLS and why it avoids overriding too many fingerprint surfaces.
- What the Linux image adds over the Windows one: a VNC viewer, a browser-based GUI and optional GPU acceleration.
- Which Kameleo plan and credentials you need to run the container, and what you can skip if you only want to try it.
In the full article: The docker run command with its flags, the Python code for profile creation and the Playwright connection, the scraping logic against the test page, and my verdict on whether Kameleo Docker is worth trying.