cdp-browser

Control a real Chrome instance via Chrome DevTools Protocol without webdriver fingerprints.

Updated Aug 27, 2026
One-click install
npx skills add https://github.com/warterbili/AUTO_REVERSE --skill cdp-browser-warterbili
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: cdp-browser
Source: https://github.com/warterbili/AUTO_REVERSE/tree/main/skills/web/cdp-browser
Command: npx skills add https://github.com/warterbili/AUTO_REVERSE --skill cdp-browser-warterbili

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires websockets, and includes scripts (resource) and references (resource) components.

What problem does it solve?

Standard browser automation tools like Selenium and Playwright leave detectable webdriver fingerprints that are immediately flagged and blocked by anti-bot systems such as PerimeterX, Cloudflare, and DataDome, preventing legitimate reverse engineering, traffic analysis, and automation of protected web properties.

Core Features & Use Cases

  • Undetectable Chrome Control: Directly control a real Chrome browser via the Chrome DevTools Protocol with no webdriver injection or automation traces, avoiding all standard bot detection signals.
  • Anti-Bot Traffic Capture: Specialized scripts to capture anti-bot SDK JavaScript code and collector API request/response pairs for reverse engineering, with per-batch isolated browser contexts to avoid session contamination.
  • General Web Automation: Perform navigation, JavaScript evaluation, element interaction, screenshot capture, and network request monitoring for any web task that requires a real browser context.
  • Use Case: A security researcher can use this skill to capture the full PerimeterX SDK and its collector POST requests from a target e-commerce site, then analyze the SDK code to reverse engineer the request signing algorithm.

Quick Start

Use the cdp-browser skill to capture anti-bot SDK code and API traffic from a target protected website, saving all collected artifacts to your workspace for reverse engineering.

Frequently Asked Questions about cdp-browser

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I control Chrome via CDP without leaving webdriver fingerprints?

You can avoid webdriver fingerprints by using direct Chrome DevTools Protocol connections to a real browser instance, bypassing standard automation tool injection that triggers anti-bot detection.

Why does Cloudflare block my Selenium scripts and how do I bypass it?

Cloudflare blocks Selenium scripts because they leave detectable webdriver fingerprints. You can bypass this by using direct CDP control of a real Chrome instance, which injects no automation traces or webdriver signals.

How do I capture anti-bot SDK JavaScript and collector API traffic for reverse engineering?

To capture anti-bot SDK JavaScript and collector API traffic, use specialized CDP scripts to intercept network requests and evaluate client-side JavaScript in isolated browser contexts, saving artifacts to your workspace for analysis.

Can I use websockets to intercept protected network requests in Chrome?

Yes, you can use the websockets dependency to establish a Chrome DevTools Protocol connection, enabling network request interception, JavaScript evaluation, and traffic monitoring without triggering bot detection systems.

What is the best way to scrape bot-protected sites without triggering DataDome?

The best way to scrape bot-protected sites without triggering DataDome is direct CDP-based browser navigation, which operates a real Chrome instance with no webdriver injection, avoiding standard bot detection signals.

Do I need isolated browser contexts to analyze anti-bot systems?

Yes, you need isolated browser contexts to analyze anti-bot systems because per-batch isolation prevents session contamination, ensuring clean traffic capture and accurate SDK artifact collection for reverse engineering.