Selenium MCP Server Guide

Quick Start: Install & Use in VS Code

Prerequisites

Node.js 18+
TypeScript
Selenium WebDriver browser drivers

1. Clone the Repository

git clone https://github.com/KiranMote/Selenium_MCP_Server.git
cd Selenium_MCP_Server

2. Install Dependencies

npm install

3. Build the Project

npm run build

4. (Optional) Test the Server

Add and run tests as needed.

5. Register the MCP Server in VS Code

Global Registration (Recommended):

Open (or create) the file:
C:\Users\\AppData\Roaming\Code\User\mcp.json

Add the following under "servers":

{
  "selenium-mcp-server": {
    "type": "stdio",
    "command": "node",
    "args": [
      "C:\\Users\\\\Selenium_MCP_Server\\dist\\index.js"
    ]
  }
}

Save the file. Now, the server is available in all VS Code workspaces.

How to use Selenium MCP server

Check in Github Copilot Chat section Selenium MCP server tools present or not.

Give Prompt As per Requirement:

It uses exploration, inspect, and more tools from that MCP and does work for us.

Features

Multi-Browser Support: Chrome, Firefox, and Edge browsers
Comprehensive Element Interaction: Click, type, get text, wait for elements
Advanced Capabilities: Screenshot capture, JavaScript execution, page navigation
Flexible Element Selection: CSS selectors, XPath, ID, class name, and tag name
Robust Error Handling: Proper validation and user-friendly error messages
TypeScript: Full type safety and modern JavaScript features

Available Tools

Browser Management: start_browser, close_browser, navigate, get_current_url, get_page_title
Element Interaction: click, type_text, get_text, wait_for_element
Advanced Features: take_screenshot, execute_script

Example Usage

{
  "tool": "start_browser",
  "arguments": {
    "browser": "chrome",
    "headless": false,
    "windowSize": "1920x1080"
  }
}

{
  "tool": "navigate",
  "arguments": {
    "url": "https://example.com"
  }
}

{
  "tool": "click",
  "arguments": {
    "selector": "#submit-button",
    "method": "css"
  }
}

{
  "tool": "type_text",
  "arguments": {
    "selector": "input[name='username']",
    "text": "user123",
    "method": "css"
  }
}

Configuration

Browser Options: browser, headless, windowSize, userAgent
Element Selection Methods: css, xpath, id, className, tagName
Wait Conditions: visible, present, clickable

Development

Building

npm run build

Running in Development

npm run dev

Project Structure

src/
  index.ts          # Main server implementation
dist/               # Compiled JavaScript output
.vscode/
  mcp.json         # MCP server configuration
.github/
  copilot-instructions.md  # Copilot development guidelines

Browser Driver Setup

Make sure you have the appropriate browser drivers installed:

Chrome: ChromeDriver
Firefox: GeckoDriver
Edge: EdgeDriver

Most drivers can be automatically managed by Selenium WebDriver 4+.

Error Handling

Input validation using Zod schemas
Browser session state checking
Timeout management for element operations
Graceful cleanup on server shutdown

Contributing

Fork the repository
Create a feature branch
Make your changes
Add tests if applicable
Submit a pull request

License

MIT License - see LICENSE file for details.

Selenium MCP Server