vibheksoni/stealth-browser-mcp

The only browser automation that bypasses anti-bot systems. AI writes network hooks, clones UIs pixel-perfect via simple chat.

What it solves

Stealth Browser MCP provides AI agents with a way to automate web browsing while avoiding detection by anti-bot systems. It allows agents to navigate through Cloudflare challenges, login walls, and other anti-automation checks that typically block standard headless browsers like Playwright.

How it works

The project implements a Model Context Protocol (MCP) server that controls real Chrome-family browser instances (Chrome, Chromium, or Microsoft Edge) using nodriver and the Chrome DevTools Protocol (CDP). By using real browser instances rather than headless drivers, it mimics human behavior more effectively to bypass bot detection.

Who it’s for

Developers building AI agents that require reliable web access to hardened sites, web scrapers who need to bypass bot detection, and users of MCP-compatible clients (like Claude Desktop or Cursor) who want to give their agents browser automation capabilities.

Highlights

  • Anti-bot resistance: Designed to pass Cloudflare and Queue-It style challenges.
  • Extensive Toolset: Offers 97 tools across 11 categories, including browser management, element interaction, and network debugging.
  • Pixel-accurate cloning: Extracts complete UI elements including CSS, DOM structure, and animations via CDP.
  • Network Inspection: Allows AI agents to inspect requests, responses, headers, and payloads.
  • Modular Architecture: Supports different modes (Full, Minimal, Xpool safe) to load only the necessary tools.
  • Dynamic Hooks: Features a restricted Python hook system to intercept or modify network traffic flows.

Related

  • Project
  • Project
  • Project
  • Project
  • Project