ClawKit Logo
ClawKitReliability Toolkit
Back to Registry
Official Verified browser automation Safety 3/5

agent-browser-stealth

Stealth browser automation with anti-detection. Launches Chromium with fingerprint randomization, webdriver flag removal, Canvas/WebGL spoofing, and permissions API masking. Use for web scraping, login automation, and session persistence on bot-protected sites. Triggers on "stealth browser", "anti-detection", "undetectable browser", "hide automation", "stealth login", "bot protection bypass".

skill-install — Terminal

Install via CLI (Recommended)

clawhub install openclaw/skills/skills/bfchain2-hub/agent-browser-stealth-xyh
Or

What This Skill Does

agent-browser-stealth is an advanced automation wrapper for Playwright designed specifically to bypass sophisticated bot-detection mechanisms on the modern web. Unlike standard headless browsers that broadcast automation signals, this skill actively modifies the browser environment to appear as a human-operated user session. It achieves this by stripping out 'navigator.webdriver' flags, injecting randomized noise into Canvas and WebGL rendering buffers to evade fingerprinting, and masking internal Chrome runtime variables. The architecture uses a CDP (Chrome DevTools Protocol) wrapper to ensure that the browser instance reports a clean environment to external security services like Cloudflare or Akamai. By providing a ref-based interaction system (e.g., e1, e2), it allows users to perform complex workflows like logins, form filling, and scraping without the instability typically associated with conventional automation scripts.

Installation

To use this skill, ensure your environment has Node.js installed. Run the following command in your terminal:

npm install -g playwright npx playwright install chromium

This will provision the necessary Chromium binary. The skill is designed to run as a standalone process through the OpenClaw agent, utilizing the 'stealth-launch.js' script to orchestrate the browser lifecycle. No additional plugins are required, as the core stealth logic is embedded within the skill's local repository structure.

Use Cases

This skill is ideal for tasks requiring reliable interactions with protected web assets. Primary use cases include: automated login and session maintenance for persistent services, data extraction from sites with aggressive anti-scraping policies, testing web applications in a production-like environment, and performing background verification tasks on platforms that detect standard bot traffic.

Example Prompts

  1. "Stealth browser: Open https://secure-portal.com and login using my saved credentials, then extract the dashboard data to a JSON file."
  2. "Use anti-detection tools to navigate to the research site, snapshot the elements, and click on the 'Export CSV' button ref."
  3. "Start a new stealth login session for the analytics platform and keep the session active so I can verify the bypass status."

Tips & Limitations

To maximize the success rate of this skill, prioritize headless-mode settings that align with your anti-detection requirements; note that while 'headless' is useful, some sophisticated sites require 'headful' mode with specific viewport settings. Always use the provided snapshot functionality to identify element references reliably, as selectors may change dynamically. Be aware that over-automation of a single endpoint can still lead to IP-based rate limiting, even if the browser fingerprint is clean. Combine this tool with rotating proxies for high-volume tasks to avoid detection at the network level.

Metadata

Stars4473
Views0
Updated2026-05-01
View Author Profile
AI Skill Finder

Not sure this is the right skill?

Describe what you want to build — we'll match you to the best skill from 16,000+ options.

Find the right skill
Add to Configuration

Paste this into your clawhub.json to enable this plugin.

{
  "plugins": {
    "official-bfchain2-hub-agent-browser-stealth-xyh": {
      "enabled": true,
      "auto_update": true
    }
  }
}

Tags(AI)

#browser-automation#stealth-scraping#bot-bypass#playwright#web-testing
Safety Score: 3/5

Flags: network-access, file-read, file-write, code-execution