AgentSkillsCN

browser-automation-pro

通过 Webhook 和 API 触发并管理 n8n 工作流。适用于需要自动化复杂多步骤流程、集成外部服务，或以编程方式触发工作流的场景。支持自托管 n8n（免费版）或 n8n 云版。

SKILL.md
--- frontmatter
name: browser-automation-pro
description: Advanced browser automation with Cloudflare bypass, stealth mode, and anti-detection. Use when regular browser/fetch fails due to bot protection, Cloudflare, or when you need to scrape protected sites. Includes puppeteer-real-browser for undetectable automation.

Browser Automation Pro

Advanced browser automation that bypasses common anti-bot protections.

When to Use

  • •Regular web_fetch returns 403/Cloudflare challenge
  • •Site has bot detection (Cloudflare, Akamai, PerimeterX)
  • •Need to interact with JavaScript-heavy SPAs
  • •Regular puppeteer gets blocked

Quick Start

Install Dependencies

bash
npm install puppeteer-real-browser puppeteer-extra puppeteer-extra-plugin-stealth

Basic Usage (Stealth Mode)

Run the stealth browser script:

bash
node skills/browser-automation-pro/scripts/stealth-fetch.js "https://example.com"

Puppeteer Real Browser (Maximum Stealth)

For sites with aggressive bot detection:

bash
node skills/browser-automation-pro/scripts/real-browser.js "https://example.com"

Available Scripts

scripts/stealth-fetch.js

Basic stealth fetch using puppeteer-extra-plugin-stealth.

  • •Bypasses most basic bot detection
  • •Faster startup than real-browser
  • •Good for: Cloudflare JS challenge, basic fingerprinting

scripts/real-browser.js

Maximum stealth using puppeteer-real-browser.

  • •Uses actual Chrome instead of Chromium
  • •Passes all fingerprint tests
  • •Good for: Aggressive bot detection, Cloudflare Turnstile

scripts/batch-fetch.js

Fetch multiple URLs with rate limiting and rotation.

  • •Configurable delays
  • •Random user agents
  • •Good for: Bulk scraping without getting blocked

Tips

  1. •Start with stealth-fetch.js - It's faster and works for 80% of cases
  2. •Use real-browser.js only when stealth-fetch fails
  3. •Add delays between requests (2-5 seconds minimum)
  4. •Rotate user agents for bulk operations

Limitations

  • •Cannot solve CAPTCHAs (need paid service like 2captcha)
  • •Very aggressive sites may still detect after many requests
  • •Slower than regular fetch (browser overhead)

References

See references/detection-methods.md for details on how bot detection works.