Browserbase

You can automate a cloud-based browser with Browserbase through an MCP server, enabling you to create and manage browser sessions, navigate pages, take screenshots, fill forms, and run JavaScript within a secure remote browser environment from your MCP client.

How to use

You interact with the Browserbase MCP Server from your MCP client by calling its tools to control a cloud browser. Start a session, navigate to pages, interact with page elements, capture screenshots, extract content, run scripts, and monitor the browser console. You can also run multiple sessions in parallel and gather content from different pages for data extraction.

How to install

Prerequisites: You need Node.js installed on your machine. You also need access to configure your MCP client to load the Browserbase MCP Server.

# Install dependencies
npm install

# Build the MCP server bundle
npm run build

Additional configuration and usage notes

Configure your MCP client to load the Browserbase MCP Server by specifying a stdio-based server that runs locally. The runtime command uses Node.js to execute the built server file and provides required environment variables for authentication.

{
  "mcpServers": {
    "browserbase": {
      "command": "node",
      "args": ["path/to/mcp-server-browserbase/browserbase/dist/index.js"],
      "env": {
        "BROWSERBASE_API_KEY": "<YOUR_BROWSERBASE_API_KEY>",
        "BROWSERBASE_PROJECT_ID": "<YOUR_BROWSERBASE_PROJECT_ID>"
      }
    }
  }
}

Notes on usage patterns

• Start with browserbase_create_session to initialize a cloud browser session. - Use browserbase_navigate to load a target URL. - Use browserbase_screenshot to capture full-page or element-specific images, customizing width, height, and a descriptive name. - Use browserbase_click and browserbase_fill to interact with page elements. - Use browserbase_evaluate to run JavaScript in the page context. - Use browserbase_get_content to extract page content, and browserbase_parallel_sessions to manage multiple sessions at once. - Access console output via console://logs and capture screenshots via screenshot://<name>.

Troubleshooting and tips

If you don’t see tools available in your client, ensure the MCP server bundle is built (npm run build), the client is configured with the correct path to dist/index.js, and the required environment variables are provided. Verify that your Node.js version is compatible with the server runtime. Check that network access to the Browserbase service is allowed if you are using API-based features.

Available tools

browserbase_create_session

Create a new cloud browser session using Browserbase. No required inputs.

browserbase_navigate

Navigate to any URL in the browser. Input: url (string).

browserbase_screenshot

Capture screenshots of the entire page or specific elements. Inputs: name (string, required), selector (string, optional), width (number, optional, default 800), height (number, optional, default 600).

browserbase_click

Click elements on the page. Input: selector (string).

browserbase_fill

Fill out input fields. Inputs: selector (string), value (string).

browserbase_evaluate

Execute JavaScript in the browser console. Input: script (string).

browserbase_get_content

Extract all content from the current page. Input: selector (string, optional).

browserbase_parallel_sessions

Create multiple browser sessions and navigate to different URLs. Input: sessions (array of {url, id}).