browser-mcp-control

Control Chrome browser tabs via MCP for AI-assisted web automation.

7|2|Updated May 16, 2026
One-click install
npx skills add https://github.com/reason-machines/mcp-skills --skill browser-mcp-control-reason-machines
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: browser-mcp-control
Source: https://github.com/reason-machines/mcp-skills/tree/main/skills/browser-mcp-control
Command: npx skills add https://github.com/reason-machines/mcp-skills --skill browser-mcp-control-reason-machines

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Automating browser tasks usually requires headless browsers that lose login sessions and trigger CAPTCHAs. This Skill connects AI assistants to your existing Chrome browser through the Browser MCP server and extension, preserving cookies and authenticated sessions while enabling programmatic navigation, clicking, typing, and data extraction. ## Core Features & Use Cases - Browser Navigation & Interaction: Navigate to URLs, click elements by CSS selector or coordinates, type into inputs, select dropdown options, and press keyboard keys. - Page Inspection: Capture ARIA accessibility tree snapshots as YAML, take full-page screenshots, and retrieve browser console logs for debugging. - Session Persistence: Operates on your existing Chrome profile via a WebSocket bridge on port 9009, with optional Redis-backed session storage. - Use Case: Automate a login flow on an internal dashboard, fill a multi-field contact form, then extract structured content from the resulting page using a snapshot for AI analysis. ## Quick Start Install the Browser MCP server and Chrome extension, then ask your AI assistant to navigate to a website and take a screenshot of the current page.

Frequently Asked Questions about browser-mcp-control

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

FAQPage Schema
How do I automate browser actions with MCP?▼

Install the Browser MCP server, register it in your MCP client config with the absolute path to dist/index.js, then install the Chrome extension and click Connect on the active tab. The AI can then call tools like browser_navigate, browser_click, and browser_type.

How to take a screenshot of a web page using Browser MCP?▼

Call the browser_screenshot tool after navigating to the target page. It accepts optional quality (0-100) and fullPage parameters and returns a base64-encoded image of the current tab.

Does Browser MCP work with logged-in sessions?▼

Yes, Browser MCP operates on your existing Chrome profile rather than a headless instance, so cookies and login sessions are preserved. This reduces CAPTCHA triggers and bot detection compared to headless automation tools.

Why is the Browser MCP extension not connecting?▼

Connection failures usually mean the extension popup was not used to click Connect on the active tab, or port 9009 is blocked or already in use. Check chrome://extensions, verify the WebSocket port with lsof -i :9009, or set a different WS_PORT.

What are the limitations of Browser MCP compared to headless automation?▼

Browser MCP requires Chrome to be open with the extension connected, supports only one active WebSocket connection at a time, and needs the tab to be active and visible to avoid timeouts. It is not suited for unattended server-side scraping.