cdp-bridge-mcp-browser-control

Connect an LLM to existing Chrome tabs for authenticated browsing tasks.

7|2|Updated May 16, 2026

Search Tags:#browser automation #mcp #cdp #screenshot #javascript execution #authenticated browsing #page scanning

One-click install

npx skills add https://github.com/Aradotso/mcp-skills --skill cdp-bridge-mcp-browser-control

Or copy as Structured Prompt for Agent▼

Please help me install this Agent Skill.
Skill: cdp-bridge-mcp-browser-control
Source: https://github.com/Aradotso/mcp-skills/tree/main/skills/cdp-bridge-mcp-browser-control
Command: npx skills add https://github.com/Aradotso/mcp-skills --skill cdp-bridge-mcp-browser-control

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It eliminates the friction of re-authenticating and re-creating browser state when an LLM needs to interact with a website, because it can directly connect to your already-open, already-logged-in Chrome tabs.

Core Features & Use Cases

Preserves session state (cookies & logins): reuse an existing browser session so authenticated sites can be scanned and acted on without transferring credentials.
LLM-friendly page understanding: scan the current page by extracting simplified HTML/text with reduced token noise.
Real interactions across navigation and tabs: list tabs, switch tabs, wait for dynamic content, execute JavaScript with CSP fallbacks, navigate, and capture screenshots.
Use case: you have an analytics dashboard open in an authenticated tab and want the agent to read specific metrics, wait for them to render, and produce a screenshot or extracted values.

Quick Start

Start the CDP Bridge MCP server, then ask the agent to list your open browser tabs using the browser_get_tabs tool.

Frequently Asked Questions about cdp-bridge-mcp-browser-control

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

FAQPage Schema

How do I let an LLM read an authenticated browser page without re-entering login credentials?▼

To use browser automation with an LLM, start the CDP Bridge MCP server and load the Chrome extension in developer mode to connect the agent to your open tabs.

Can I execute JavaScript on a page using an LLM agent?▼

Yes, you can capture screenshots and retrieve cookies from authenticated tabs using the browser_screenshot and browser_cookies tools provided by the MCP server.

What is the best way to extract content from dynamic web pages with an LLM?▼

You need a Chrome extension loaded in developer mode and a running CDP Bridge MCP server exposing functional tool execution endpoints for tasks like browser_scan and browser_execute_js.

How do I manage multiple browser tabs during LLM browser automation?▼

If JavaScript execution fails on a page, the CDP bridge applies CSP-aware fallbacks to bypass Content Security Policy restrictions and successfully run the required scripts.