What WebBrowser is your OpenClaw/dBot using? · OpenClawBuilders/AI Automation

Jeffrey Thompson

2d • General discussion

What WebBrowser is your OpenClaw/dBot using?

Top 10 Browsers for Headless Ubuntu 22.04 (Bot/Automation Use)

Ranked for automation, Playwright compatibility, and minimal install environments:

#1. Chromium — Best overall. Built-in headless mode, best CDP support, Playwright default.

#2. Google Chrome (stable) — Better site compatibility on some auth flows vs Chromium.

#3. Firefox — Strong Playwright/Selenium support, more stable headless in edge cases.

#4. Brave is Chromium-based, so it works with CDP and Playwright — you just have to point Playwright

#5. Playwright's bundled Chromium — Auto-managed, versioned, zero config. Best if already using

Playwright.

#6. Playwright's bundled Firefox — Same benefit, Firefox engine.

#7. Playwright's bundled WebKit — Lightweight, good for simple scraping.

#8. Chrome for Testing — Google's pinned automation-specific build. No auto-updates breaking things.

#9. ungoogled-chromium — Privacy-stripped Chromium, useful for avoiding bot detection.

#10. Lynx — Text-only, ultra-minimal. Only useful for plain HTML scraping, no JS.

#11. PhantomJS — Deprecated/dead but still runs legacy scripts. Avoid for new builds.

---

Watch Out For:

--no-sandbox is required in Docker but is a security risk on shared hosts

Docker default shared memory (64MB) crashes Chromium under load — add --shm-size=1gb

Upgrading Playwright pins new browser versions — re-run playwright install after upgrades

Each Chromium instance uses 200-400MB RAM — plan accordingly for multi-bot setups

Standard headless Chromium is easily fingerprinted — use playwright-stealth if hitting bot-detection sites

---

Brave belongs on the list.

Why it got left off: Brave doesn't have native Playwright/Selenium support out of the box — it requires manual path configuration to use as a Chromium substitute. So it got filtered out mentally as "extra steps." That's a valid tradeoff to know, not a reason to ignore it.

Where Brave fits:

Brave is Chromium-based, so it works with CDP and Playwright — you just have to point Playwright at the Brave binary manually:

pythonbrowser = playwright.chromium.launch(executable_path="/usr/bin/brave-browser")

Reasons to use it over standard Chromium:

Built-in ad/tracker blocking — reduces page noise and speeds up loads

Better bot detection evasion out of the box vs vanilla Chromium

Fingerprinting protections baked in (useful for Polymarket-type sites)

Shields can be toggled off per-session if they interfere with scraping

Reasons to skip it:

Not in Ubuntu's default apt repos — requires manual PPA setup

Brave updates can shift the binary path and break your launch config

Shields occasionally block content you actually want to scrape