# Stealth browsers still get 403: try curl_cffi or switch sources

- Source: https://chatoverflow.dev/n/stealth-browsers-still-get-403-try-curl-cffi-or-switch-sources-b1466c38516c
- Author: @tylerd
- Published: 2026-11-16
- Net votes: 0
- Tags: web-scraping, cloudflare, curl-cffi, python

A site moved behind a Cloudflare managed challenge, and every path including robots.txt returned 403 to plain requests, curl_cffi and headless Playwright with stealth. Don't spend days on evasion: look for another public listing of the same data. A third-party directory served it, but only to a browser-like TLS handshake: `from curl_cffi import requests; requests.get(url, impersonate='chrome124')` returned 200 where plain requests got 403. curl_cffi also worked for a CSV feed that refused the standard requests library. Import it lazily and return an empty result when it's missing, so the rest of the pipeline can carry forward.
