Skip to content
MCP server

Selenium MCP

MCP server giving AI assistants Selenium WebDriver browser automation tools

Runs locallyStatus and history
Install
Pricing
Open source
Platforms
CLI, vscode, macOS
Runs
Locally
Version
0.4.6
Listing updated
today

Similar MCP servers

About Selenium MCP

Selenium MCP is an open-source MCP server for browser automation. It runs locally on your machine and is started by your AI client.

The README describes this as a bridge between MCP clients and Selenium WebDriver: it exposes WebDriver actions as MCP tools so an assistant can open a browser, move through pages, interact with elements and read results. The tool catalogue is organised into browser management, cookies, windows, frames, forms, scrolling and element state checks, covering roughly sixty named tools.

It is written in TypeScript and runs over stdio, typically launched with npx from the selenium-webdriver-mcp npm package. The README says Chrome, Firefox, Safari and Edge are supported, with WebDrivers managed by selenium-webdriver, and requires Node.js 18 or later.

The server suits developers and QA engineers who want an assistant to drive a real browser for scraping, testing or generating test code. The README's example prompt has an agent register on a demo banking site and then produce page-object tests.

Install

Run in a terminal

claude mcp add selenium-mcp --env YOUR_API_KEY='<YOUR_API_KEY>' -- npx -y selenium-webdriver-mcp

Add to claude_desktop_config.json

{
  "mcpServers": {
    "selenium-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "selenium-webdriver-mcp"
      ],
      "env": {
        "YOUR_API_KEY": "<YOUR_API_KEY>"
      }
    }
  }
}

Add to ~/.cursor/mcp.json

{
  "mcpServers": {
    "selenium-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "selenium-webdriver-mcp"
      ],
      "env": {
        "YOUR_API_KEY": "<YOUR_API_KEY>"
      }
    }
  }
}

Add to .vscode/mcp.json

{
  "servers": {
    "selenium-mcp": {
      "type": "stdio",
      "command": "npx",
      "args": [
        "-y",
        "selenium-webdriver-mcp"
      ],
      "env": {
        "YOUR_API_KEY": "<YOUR_API_KEY>"
      }
    }
  }
}

Add to ~/.codeium/windsurf/mcp_config.json

{
  "mcpServers": {
    "selenium-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "selenium-webdriver-mcp"
      ],
      "env": {
        "YOUR_API_KEY": "<YOUR_API_KEY>"
      }
    }
  }
}

Add to your client's MCP settings (most use this shape)

{
  "mcpServers": {
    "selenium-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "selenium-webdriver-mcp"
      ],
      "env": {
        "YOUR_API_KEY": "<YOUR_API_KEY>"
      }
    }
  }
}

Using several servers? Build one config file with the MCP config generator.

Configuration

  • YOUR_API_KEYYour API key for the service

Tools this server provides (9)

Taken from the project documentation; the running server may differ.

browser_open
Open a new browser session
browser_navigate
Navigate to a URL
browser_click
Click on an element
browser_type
Type text into an element
browser_find_element
Find an element on the page
browser_screenshot
Take a screenshot
browser_execute_script
Execute JavaScript code
browser_get_cookies
Get all cookies from the current browser session
browser_close
Close current browser session

What it does

Multi-browser automation
Drives Chrome, Firefox, Safari and Edge through Selenium WebDriver, according to the README.
Broad element interaction
Tools cover clicking, typing, hovering, drag and drop, dropdowns, checkboxes and file uploads.
Window and frame handling
Switches between browser windows by handle, title, URL or index, and moves in and out of iframes.
Cookie management
Lists, reads, sets and deletes browser cookies through dedicated tools.
Screenshots and JavaScript
Captures screenshots and runs JavaScript in the page context.
Typed with Zod validation
Written in TypeScript and validates tool inputs with Zod, per the README.

Good for

  • Letting an AI assistant walk through a web app to check a user flow
  • Generating Selenium test code from a browser session an agent has explored
  • Scraping page content and element text for data extraction
  • Taking screenshots of pages during automated checks

Pricing

Open source

Free and MIT licensed; run locally via npm.

Selenium MCP FAQ

How is the server started?

The README's standard configuration runs npx -y selenium-webdriver-mcp@latest as a stdio command in MCP clients such as Claude Desktop and VS Code.

What license does it use?

The README states the project is released under the MIT License.

What does it require to run?

The README lists Node.js 18.0.0 or higher, an installed Chrome, Firefox, Safari or Edge browser, and Windows, macOS or Linux.

Is Selenium MCP free?

Yes. Selenium MCP is open source, so you can use and self-host it for free.

How do I add Selenium MCP to Claude Code?

Run this command in your terminal: claude mcp add selenium-mcp --env YOUR_API_KEY='<YOUR_API_KEY>' -- npx -y selenium-webdriver-mcp

What are the best alternatives to Selenium MCP?

The closest alternatives to Selenium MCP are Selenium Webdriver MCP, Chrome DevTools MCP, MCP Browser, MCP Browser Extension and Turbo-Browser-MCP.

Where this comes from

  • Description: written from the project's README and its MCP Registry entry.
  • Install steps: from the official MCP Registry entry.

How AIPoweredBot checks servers