Browserbase MCP

v2.2.0Browser Automationstable

This server provides cloud browser automation capabilities using Browserbase, Puppeteer, and Stagehand. This server enables LLMs to interact with web pages, take screenshots, and execute JavaScript in a cloud browser environment.

aibrowserchromechromiummcp
Share:
3,348
Stars
0
Downloads
0
Weekly
0/5

What is Browserbase MCP?

Browserbase MCP is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to this server provides cloud browser automation capabilities using browserbase, puppeteer, and stagehand. this server enables llms to interact with web pages, take screenshots, and execute javascript in...

This server provides cloud browser automation capabilities using Browserbase, Puppeteer, and Stagehand. This server enables LLMs to interact with web pages, take screenshots, and execute JavaScript in a cloud browser environment.

This server falls under the Browser Automation category on MCPgee, the world's largest MCP server directory with 33,000+ servers.

Features

  • browserbase_session_create
  • browserbase_session_close
  • browserbase_stagehand_navigate
  • browserbase_stagehand_act
  • browserbase_stagehand_extract

Use Cases

Cloud browser automation for LLMs
Screenshot capture and JavaScript execution
browserbase

Maintainer

LicenseMIT License
Languagetypescript
Versionv2.2.0
UpdatedMay 21, 2026
Statushealthy
Maintenanceactive

Works with

ClaudeOpenAIwindowsmacoslinux

Installation

NPM

npx -y @browserbasehq/mcp-server-browserbase

Manual Installation

npx -y @browserbasehq/mcp-server-browserbase

Configuration

Configuration Details

Config File

claude_desktop_config.json

Performance

Response Metrics

Response Time< 200ms
ThroughputMedium

Resource Usage

Memory UsageLow
CPU UsageLow

How to Set Up and Use Browserbase MCP

The Browserbase MCP server gives AI assistants full cloud browser automation capabilities by connecting them to Browserbase's remote Chrome infrastructure via the Stagehand framework. LLMs can navigate to URLs, observe interactive elements on a page, perform click and form actions, extract structured data, and take screenshots — all without running a local browser. This makes it well-suited for research automation, web scraping, and agentic workflows that require real browser rendering rather than raw HTTP fetches.

Prerequisites

  • Node.js 18+ to run the server via npx
  • A Browserbase account with a valid API key (sign up at browserbase.com)
  • A Browserbase Project ID from your dashboard
  • A Gemini API key (default model) or an alternative model API key if overriding the default
  • An MCP-compatible client such as Claude Desktop
1

Create a Browserbase account and project

Sign up at browserbase.com, create a new project, and copy your API key and Project ID from the dashboard. These credentials are required for every browser session.

2

Test the server locally

Run the server directly with npx to confirm your credentials work before adding it to a client config.

BROWSERBASE_API_KEY=your_api_key \
BROWSERBASE_PROJECT_ID=your_project_id \
npx -y @browserbasehq/mcp-server-browserbase
3

Add the server to Claude Desktop configuration

Edit your claude_desktop_config.json to include the Browserbase server with your credentials passed as environment variables.

{
  "mcpServers": {
    "browserbase": {
      "command": "npx",
      "args": ["-y", "@browserbasehq/mcp-server-browserbase"],
      "env": {
        "BROWSERBASE_API_KEY": "your_api_key",
        "BROWSERBASE_PROJECT_ID": "your_project_id",
        "GEMINI_API_KEY": "your_gemini_api_key"
      }
    }
  }
}
4

Restart Claude Desktop

Fully quit and relaunch Claude Desktop so it connects to the Browserbase MCP server. You should see it listed in the available tools.

5

Start a browser session and navigate

Once connected, ask Claude to open a browser session and navigate to a URL. The server will create a Browserbase cloud session and return the page content.

Browserbase MCP Examples

Client configuration

Full claude_desktop_config.json block for the Browserbase MCP server with all required environment variables.

{
  "mcpServers": {
    "browserbase": {
      "command": "npx",
      "args": ["-y", "@browserbasehq/mcp-server-browserbase"],
      "env": {
        "BROWSERBASE_API_KEY": "your_browserbase_api_key",
        "BROWSERBASE_PROJECT_ID": "your_browserbase_project_id",
        "GEMINI_API_KEY": "your_gemini_api_key"
      }
    }
  }
}

Prompts to try

These prompts demonstrate the browser automation tools the server exposes.

- "Open a browser session and navigate to https://news.ycombinator.com, then list the top 5 story titles"
- "Go to https://example.com/login, fill in the email field with [email protected], and click the Sign In button"
- "Extract all product names and prices from https://books.toscrape.com"
- "Take a screenshot of https://anthropic.com and describe what you see"
- "Observe all clickable elements on the current page and tell me what actions are available"

Troubleshooting Browserbase MCP

Session creation fails with 401 Unauthorized

Double-check that BROWSERBASE_API_KEY and BROWSERBASE_PROJECT_ID are both set correctly. The Project ID is a UUID found in your Browserbase dashboard under project settings, not the project name.

The observe or act tools return empty results or fail to find elements

Some pages require JavaScript to fully render. Try adding a brief wait instruction in your prompt ('wait for the page to fully load before extracting'). If the page uses anti-bot measures, enable proxies by passing --proxies in the server args.

GEMINI_API_KEY errors even though you want to use a different model

Pass --modelName and --modelApiKey as additional args to override the default Gemini model: add "--modelName", "claude-3-5-sonnet-20241022", "--modelApiKey", "your_anthropic_key" to the args array.

Frequently Asked Questions about Browserbase MCP

What is Browserbase MCP?

Browserbase MCP is a Model Context Protocol (MCP) server that this server provides cloud browser automation capabilities using browserbase, puppeteer, and stagehand. this server enables llms to interact with web pages, take screenshots, and execute javascript in a cloud browser environment. It connects AI assistants to external tools and data sources through a standardized interface.

How do I install Browserbase MCP?

Install via npm with the command: npx -y @browserbasehq/mcp-server-browserbase. Then add the server configuration to your AI client's JSON config file (e.g., claude_desktop_config.json or .cursor/mcp.json).

Which AI clients work with Browserbase MCP?

Browserbase MCP works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.

Is Browserbase MCP free to use?

Yes, Browserbase MCP is open source and available under the MIT License license. You can use it freely in both personal and commercial projects.

Browse More Browser Automation MCP Servers

Explore all browser automation servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.

Quick Config Preview

{ "mcpServers": { "browserbase": { "command": "npx", "args": ["-y", "@browserbasehq/mcp-server-browserbase"] } } }

Add this to your claude_desktop_config.json or .cursor/mcp.json

Read the full setup guide →

Ready to use Browserbase MCP?

Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.

33,000+ ServersFree & Open SourceStep-by-Step Guides