Playwright MCP Automation Tool

A generic, AI-powered automation testing tool that allows teams to define web app test scenarios in natural language or structured JSON and execute them using the Model Context Protocol.

Features

• Natural Language Selectors: Click buttons and fill fields using their labels, text, or roles (e.g., "Click Login" or "Fill 'Email' with '[email protected]'").
• Self-Healing Capability: When used with an LLM, the agent can inspect the page state and find alternative ways to complete a step if a selector fails.
• Generic Scenarios: Easily define multiple scenarios with assertions for any web application.
• Screenshots: Automatic or manual full-page screenshots for debugging and reporting.

Setup

1. Install Dependencies:

   npm install
   npx playwright install chromium

2. Build the Server:

   npm run build

3. Configure with your LLM (e.g., Claude Desktop): Add the following to your claude_desktop_config.json:

   {
     "mcpServers": {
       "playwright-automation": {
         "command": "node",
         "args": ["/path/to/mcp-playwright-automation/dist/index.js"]
       }
     }
   }

Usage for Organizations

Dev teams can provide a test configuration file like this:

Example test-config.json

{
  "baseUrl": "https://example.com",
  "scenarios": [
    {
      "name": "User Login Flow",
      "steps": [
        "Navigate to /login",
        "Fill field 'Username' with 'testuser'",
        "Fill field 'Password' with 'password123'",
        "Click the 'Sign In' button"
      ],
      "assertions": [
        "URL should contain '/dashboard'",
        "Page should show text 'Welcome back'"
      ]
    }
  ]
}

Running Tests

Once the MCP server is connected to your AI agent, simply give it the configuration file and say:

"Run these automation scenarios using the playwright-automation tool and provide a report."

The AI will:

1. Initialize the session.
2. Execute each step sequentially.
3. Verify assertions.
4. Take screenshots if errors occur.
5. Summarize the results.