mcp-browser-server

mcp-browser-server

A MCP server that gives AI assistants full browser control, enabling navigation, clicking, form filling, and screenshots via any MCP client.

Category
Visit Server

README

mcp-browser-server

Give your AI assistant a real browser. Navigate, click, fill forms, and take screenshots via any MCP client.

License: MIT MCP

A Model Context Protocol (MCP) server that gives AI assistants like Claude full browser control. Connect to local or cloud browsers (AnchorBrowser, Browserbase) and automate web interactions directly from your AI workflows.

What it does

Instead of telling your AI agent how to scrape a website, you give it a real browser. The AI can:

  • Navigate to any URL
  • Click buttons and links
  • Fill out and submit forms
  • Take screenshots
  • Read page content and extract data
  • Handle authentication (SSO, multi-step logins)

MCP Tools Provided

Tool Description
browser_navigate Go to a URL
browser_click Click an element by selector or description
browser_type Type text into an input field
browser_screenshot Take a screenshot of the current page
browser_get_content Get page HTML or extracted text
browser_evaluate Run JavaScript in the browser
browser_wait Wait for an element or condition
browser_scroll Scroll the page

Installation

npm install -g mcp-browser-server

Or clone and run locally:

git clone https://github.com/mehranakila56-ops/mcp-browser-server.git
cd mcp-browser-server
npm install

Configuration

With Claude Desktop

Add to ~/Library/Application Support/Claude/claude_desktop_config.json:

{
  "mcpServers": {
    "browser": {
      "command": "npx",
      "args": ["mcp-browser-server"],
      "env": {
        "BROWSER_PROVIDER": "local",
        "HEADLESS": "true"
      }
    }
  }
}

With Cloud Browser (AnchorBrowser)

For stealth mode, CAPTCHA solving, and residential proxies:

{
  "mcpServers": {
    "browser": {
      "command": "npx",
      "args": ["mcp-browser-server"],
      "env": {
        "BROWSER_PROVIDER": "anchorbrowser",
        "ANCHOR_API_KEY": "your-api-key-here"
      }
    }
  }
}

Environment Variables

Variable Default Description
BROWSER_PROVIDER local local, anchorbrowser, browserbase
HEADLESS true Run browser in headless mode (local only)
ANCHOR_API_KEY AnchorBrowser API key
BROWSERBASE_API_KEY Browserbase API key
SESSION_TIMEOUT_MS 300000 Browser session timeout (5 min)
SCREENSHOT_FORMAT png Screenshot format: png or jpeg

Usage Examples

Once configured with Claude, you can say:

"Go to github.com/trending and show me the top 5 repos today"

"Log into our staging environment at staging.example.com and check if the signup form works"

"Take a screenshot of https://example.com and describe what you see"

"Fill out the contact form at example.com/contact with my info: name=Jane, email=jane@example.com, message='Hello'"

Architecture

Claude / AI Assistant
       ↓ MCP protocol
mcp-browser-server
       ↓ WebSocket / CDP
Browser (local Chromium or cloud)
       ↓
Target Website

Why Cloud Browsers?

Local browsers get blocked by CAPTCHAs and bot detection. Cloud browsers like AnchorBrowser provide:

  • ✅ Human fingerprinting (passes Cloudflare, DataDome)
  • ✅ Residential proxy rotation
  • ✅ CAPTCHA auto-solving
  • ✅ Persistent sessions across runs
  • ✅ No local Chrome installation needed

Development

npm install
npm run dev   # Starts MCP server with hot reload

# Run against a specific provider
BROWSER_PROVIDER=local npm run dev

Contributing

PRs welcome! Please open an issue first for significant changes.

License

MIT


Related: awesome-ai-browser-agents — curated list of browser automation tools for AI agents

Recommended Servers

playwright-mcp

playwright-mcp

A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.

Official
Featured
TypeScript
Magic Component Platform (MCP)

Magic Component Platform (MCP)

An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.

Official
Featured
Local
TypeScript
Audiense Insights MCP Server

Audiense Insights MCP Server

Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.

Official
Featured
Local
TypeScript
VeyraX MCP

VeyraX MCP

Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.

Official
Featured
Local
graphlit-mcp-server

graphlit-mcp-server

The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.

Official
Featured
TypeScript
Kagi MCP Server

Kagi MCP Server

An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.

Official
Featured
Python
E2B

E2B

Using MCP to run code via e2b.

Official
Featured
Neon Database

Neon Database

MCP server for interacting with Neon Management API and databases

Official
Featured
Exa Search

Exa Search

A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.

Official
Featured
Qdrant Server

Qdrant Server

This repository is an example of how to create a MCP server for Qdrant, a vector search engine.

Official
Featured