### Getting Started Prompts

Source: https://github.com/executeautomation/mcp-playwright/blob/main/docs/docs/playwright-web/Resize-Prompts-Guide.md

Commonly used prompts for initiating basic, responsive, and orientation-based tests.

```text
Navigate to [your-site.com] and test on iPhone 13
```

```text
Test [your-site.com] on iPhone 13, iPad Pro 11, and Desktop Chrome
```

```text
Test [your-site.com] on iPhone 13 in both portrait and landscape
```

--------------------------------

### Install Dependencies

Source: https://github.com/executeautomation/mcp-playwright/blob/main/docs/docs/local-setup/Installation.mdx

Navigate to the project directory and install required packages.

```bash
cd mcp-playwright
npm install
```

--------------------------------

### Start Playwright MCP Server

Source: https://github.com/executeautomation/mcp-playwright/blob/main/examples/http-mode-example.md

Use npx or npm to install and start the Playwright MCP Server on a specified port. Ensure the port is available before starting.

```bash
# Using npx
npx @executeautomation/playwright-mcp-server --port 8931
```

```bash
# Or after global installation
npm install -g @executeautomation/playwright-mcp-server
playwright-mcp-server --port 8931
```

--------------------------------

### Install Dependencies and Build Project

Source: https://github.com/executeautomation/mcp-playwright/blob/main/DOCKER.md

Install production dependencies and build the project before creating the Docker image to ensure a minimal image size.

```bash
npm install --omit=dev
npm run build
```

--------------------------------

### Start the HTTP Server

Source: https://github.com/executeautomation/mcp-playwright/blob/main/docs/docs/playwright-web/HTTP-SSE-Transport.mdx

Commands to launch the Playwright MCP server in HTTP mode using npx or global installation.

```bash
npx @executeautomation/playwright-mcp-server --port 8931
```

```bash
# Install globally
npm install -g @executeautomation/playwright-mcp-server

# Run the server
playwright-mcp-server --port 8931
```

```bash
# Option 1: Update to latest version
npx @executeautomation/playwright-mcp-server@latest --port 8931

# Option 2: Use global installation instead
npm install -g @executeautomation/playwright-mcp-server
playwright-mcp-server --port 8931
```

--------------------------------

### Run MCP Playwright Server with Automatic Browser Installation

Source: https://github.com/executeautomation/mcp-playwright/blob/main/docs/docs/release.mdx

Execute the server command to automatically detect and install missing browsers on first use. No manual browser installation is required.

```bash
npx @executeautomation/playwright-mcp-server
```

--------------------------------

### Build Project

Source: https://github.com/executeautomation/mcp-playwright/blob/main/docs/docs/local-setup/Installation.mdx

Install TypeScript and compile the project for local use.

```bash
npm install --save-dev typescript
npm run build
npm link
```

--------------------------------

### Navigation and Resize Execution

Source: https://github.com/executeautomation/mcp-playwright/blob/main/docs/docs/playwright-web/Resize-Prompts-Guide.md

Example of combining navigation and resizing commands.

```text
Navigate to https://example.com and test it on iPhone 13
```

```javascript
await playwright_navigate({ url: "https://example.com" });
await playwright_resize({ device: "iPhone 13" });
```

--------------------------------

### Install and Run MCP Inspector

Source: https://github.com/executeautomation/mcp-playwright/blob/main/docs/docs/local-setup/Debugging.mdx

Use this command to launch the inspector with your server's entry point.

```bash
npx @modelcontextprotocol/inspector node dist/index.js
```

--------------------------------

### Test npx and Global Installation

Source: https://github.com/executeautomation/mcp-playwright/blob/main/docs/docs/playwright-web/HTTP-SSE-Transport.mdx

Commands to test both npx and global installation methods for the MCP Playwright server. It's recommended to try npx first.

```bash
# Try npx first
npx @executeautomation/playwright-mcp-server@latest --port 8931

# If no output, use global install
npm install -g @executeautomation/playwright-mcp-server
playwright-mcp-server --port 8931
```

--------------------------------

### Log message structure examples

Source: https://github.com/executeautomation/mcp-playwright/blob/main/docs/docs/playwright-web/HTTP-SSE-Transport.mdx

Example JSON log entries for client connection, message transmission, and disconnection events.

```json
// Client connects
{"level":"info","message":"Incoming request","context":{"method":"GET","path":"/mcp"}}
{"level":"info","message":"SSE connection request received","context":{"endpoint":"/mcp"}}
{"level":"info","message":"Transport registered","context":{"sessionId":"...","activeTransports":1}}

// Client sends message
{"level":"info","message":"Incoming request","context":{"method":"POST","path":"/mcp"}}
{"level":"POST message received","context":{"sessionId":"...","availableTransports":[...]}}

// Client disconnects
{"level":"info","message":"SSE connection closed","context":{"sessionId":"...","endpoint":"/mcp"}}
{"level":"info","message":"Transport unregistered","context":{"activeTransports":0}}
```

--------------------------------

### Configure Browser Environment

Source: https://github.com/executeautomation/mcp-playwright/blob/main/docs/docs/playwright-web/HTTP-SSE-Transport.mdx

Install required browsers and set display variables for headless server environments.

```bash
npx playwright install chromium

export DISPLAY=:99
```

--------------------------------

### Testing Multiple Devices

Source: https://github.com/executeautomation/mcp-playwright/blob/main/docs/docs/playwright-web/Device-Quick-Reference.md

Example of iterating through multiple device presets and capturing screenshots for each.

```javascript
await playwright_navigate({ url: "homepage" });

await playwright_resize({ device: "iPhone 13" });
await playwright_screenshot({ name: "mobile" });

await playwright_resize({ device: "iPad Pro 11" });
await playwright_screenshot({ name: "tablet" });

await playwright_resize({ device: "Desktop Chrome" });
await playwright_screenshot({ name: "desktop" });
```

--------------------------------

### Start Server with Docker Compose

Source: https://github.com/executeautomation/mcp-playwright/blob/main/DOCKER.md

Start the Playwright MCP server using Docker Compose. This command assumes a `docker-compose.yml` file is present.

```bash
docker compose run --rm playwright-mcp
```

--------------------------------

### Install Playwright MCP Server via Smithery

Source: https://github.com/executeautomation/mcp-playwright/blob/main/docs/docs/intro.mdx

Installs the Playwright MCP Server package for Claude Desktop using the Smithery CLI. Ensure you have npx installed.

```bash
npx @smithery/cli install @executeautomation/playwright-mcp-server --client claude
```

--------------------------------

### Install Playwright MCP Server via MCP-GET

Source: https://github.com/executeautomation/mcp-playwright/blob/main/docs/docs/intro.mdx

Installs the Playwright MCP Server package using the mcp-get tool. This is a convenient way to manage MCP installations.

```bash
npx @michaellatman/mcp-get@latest install @executeautomation/playwright-mcp-server
```

--------------------------------

### Server Output and Configuration

Source: https://github.com/executeautomation/mcp-playwright/blob/main/docs/docs/playwright-web/HTTP-SSE-Transport.mdx

Example console output showing server status, endpoints, and the required client configuration JSON.

```text
⏳ Initializing Playwright MCP Server on port 8931...

🚀 Starting Playwright MCP Server (HTTP Mode)...

==============================================
Playwright MCP Server (HTTP Mode)
==============================================
Listening: 127.0.0.1:8931 (localhost only)
Version: 1.0.9

SECURITY: Server is bound to localhost only.
          Not accessible from external networks.

ENDPOINTS:
- SSE Stream:     GET  http://localhost:8931/sse
- Messages:       POST http://localhost:8931/messages?sessionId=<id>
- MCP (unified):  GET  http://localhost:8931/mcp
- MCP (unified):  POST http://localhost:8931/mcp?sessionId=<id>
- Health Check:   GET  http://localhost:8931/health

CLIENT CONFIGURATION:
{
  "mcpServers": {
    "playwright": {
      "url": "http://localhost:8931/mcp",
      "type": "http"
    }
  }
}
==============================================

Monitoring HTTP server listening on port 55123
Health check: http://localhost:55123/health
Metrics: http://localhost:55123/metrics
```

--------------------------------

### Orientation Testing

Source: https://github.com/executeautomation/mcp-playwright/blob/main/docs/docs/playwright-web/Device-Quick-Reference.md

Example of switching a device to landscape orientation.

```javascript
await playwright_resize({ device: "iPhone 13" });
// Test portrait menu

await playwright_resize({ 
  device: "iPhone 13", 
  orientation: "landscape" 
});
// Test landscape menu
```

--------------------------------

### Testing a Single Device

Source: https://github.com/executeautomation/mcp-playwright/blob/main/docs/docs/playwright-web/Device-Quick-Reference.md

Example of navigating to a URL and resizing the viewport to a specific device preset.

```javascript
await playwright_navigate({ url: "my-site.com" });
await playwright_resize({ device: "iPhone 13" });
```

--------------------------------

### Execute Build Script

Source: https://github.com/executeautomation/mcp-playwright/blob/main/DOCKER.md

Run the provided build script to install dependencies and build the project for Docker.

```bash
chmod +x docker-build.sh
./docker-build.sh
```

--------------------------------

### Configure VS Code GitHub Copilot

Source: https://github.com/executeautomation/mcp-playwright/blob/main/docs/docs/playwright-web/HTTP-SSE-Transport.mdx

Steps to start the server and update VS Code settings for HTTP transport.

```bash
playwright-mcp-server --port 8931
```

```json
{
  "github.copilot.chat.mcp.servers": {
    "playwright": {
      "url": "http://localhost:8931/mcp",
      "type": "http"
    }
  }
}
```

--------------------------------

### Install Playwright Browsers Manually

Source: https://github.com/executeautomation/mcp-playwright/blob/main/README.md

Manually install Playwright browser binaries if automatic installation fails or if you prefer manual control. You can install all browsers or specific ones like Chromium, Firefox, or WebKit.

```bash
npx playwright install
```

```bash
npx playwright install chromium
```

```bash
npx playwright install firefox
```

```bash
npx playwright install webkit
```

--------------------------------

### Install Playwright MCP Server in VS Code CLI

Source: https://github.com/executeautomation/mcp-playwright/blob/main/README.md

Install the Playwright MCP server in VS Code using the command-line interface.

```bash
code --add-mcp '{"name":"playwright","command":"npx","args":["@executeautomation/playwright-mcp-server"]}'
```

--------------------------------

### Build and Run with Docker Compose

Source: https://github.com/executeautomation/mcp-playwright/blob/main/DOCKER.md

Build the Docker image and then start the Playwright MCP server using Docker Compose.

```bash
docker compose build
docker compose run --rm playwright-mcp
```

--------------------------------

### Progressive Testing Workflow

Source: https://github.com/executeautomation/mcp-playwright/blob/main/docs/docs/playwright-web/Resize-Prompts-Guide.md

Example of a multi-step responsive testing prompt and the corresponding AI execution code.

```text
Test this page's responsiveness:
1. Start with mobile (iPhone 13)
2. Then tablet (iPad Pro 11)
3. Finally desktop (Desktop Chrome)

Take screenshots at each size and check if the layout looks correct.
```

```javascript
// The AI executes:
await playwright_navigate({ url: "https://example.com" });
await playwright_resize({ device: "iPhone 13" });
await playwright_screenshot({ name: "mobile-view" });

await playwright_resize({ device: "iPad Pro 11" });
await playwright_screenshot({ name: "tablet-view" });

await playwright_resize({ device: "Desktop Chrome" });
await playwright_screenshot({ name: "desktop-view" });
```

--------------------------------

### Perform HTTP GET Request

Source: https://context7.com/executeautomation/mcp-playwright/llms.txt

Executes a GET request with support for authentication tokens and custom headers.

```javascript
// Simple GET request
await playwright_get({
  url: "https://api.example.com/users"
});

// GET with Bearer token
await playwright_get({
  url: "https://api.example.com/profile",
  token: "eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9..."
});

// GET with custom headers
await playwright_get({
  url: "https://api.example.com/data",
  headers: {
    "X-API-Key": "your-api-key",
    "Accept": "application/json"
  }
});
```

--------------------------------

### Custom Dimensions

Source: https://github.com/executeautomation/mcp-playwright/blob/main/docs/docs/playwright-web/Device-Quick-Reference.md

Example of setting a viewport size using specific width and height values.

```javascript
await playwright_resize({ width: 1024, height: 768 });
```

--------------------------------

### Install Playwright MCP Server via NPM

Source: https://github.com/executeautomation/mcp-playwright/blob/main/docs/docs/intro.mdx

Installs the Playwright MCP Server package globally using npm. This command can be run in any integrated terminal.

```bash
npm install -g @executeautomation/playwright-mcp-server
```

--------------------------------

### Start Codegen Session

Source: https://context7.com/executeautomation/mcp-playwright/llms.txt

Initializes a recording session to capture Playwright actions and generate test files.

```javascript
// Start recording session
await start_codegen_session({
  options: {
    outputPath: "/path/to/tests",
    testNamePrefix: "LoginTest",
    includeComments: true
  }
});

// Returns: { sessionId: "uuid", options: {...}, message: "Started codegen session..." }
```

--------------------------------

### start_codegen_session

Source: https://context7.com/executeautomation/mcp-playwright/llms.txt

Starts a session to record Playwright actions and generate test files.

```APIDOC
## start_codegen_session

### Description
Start a code generation session to record Playwright actions and generate test files.

### Parameters
#### Request Body
- **options** (object) - Required - Configuration for the session (outputPath, testNamePrefix, includeComments).

### Response
#### Success Response (200)
- **sessionId** (string) - The unique ID of the session.
- **options** (object) - The configuration used.
- **message** (string) - Status message.
```

--------------------------------

### Debugging Device-Specific Issues

Source: https://github.com/executeautomation/mcp-playwright/blob/main/docs/docs/playwright-web/Device-Quick-Reference.md

Example of running a test sequence across multiple device sizes to isolate responsive issues.

```javascript
await playwright_resize({ device: "iPhone SE" });
// Check button

await playwright_resize({ device: "iPhone 13" });
// Check button

await playwright_resize({ device: "iPhone 15 Pro Max" });
// Check button
```

--------------------------------

### Start Playwright MCP Server in HTTP Mode

Source: https://github.com/executeautomation/mcp-playwright/blob/main/README.md

Start the Playwright MCP server as a standalone HTTP server on a specified port. This is useful for running headed browsers on systems without a display or from worker processes.

```bash
# Using npx
npx @executeautomation/playwright-mcp-server --port 8931

# Or after global installation
playwright-mcp-server --port 8931
```

--------------------------------

### Manually Build Docker Image

Source: https://github.com/executeautomation/mcp-playwright/blob/main/DOCKER.md

Build the Docker image for the Playwright MCP Server locally. Ensure dependencies are installed with `--omit=dev`.

```bash
docker build -t mcp-playwright:latest .
```

--------------------------------

### Custom Dockerfile for Browser Installation

Source: https://github.com/executeautomation/mcp-playwright/blob/main/DOCKER.md

Create a custom Dockerfile to pre-install Playwright browsers within the Docker image, which can be useful if the default image skips browser downloads.

```dockerfile
FROM mcp-playwright:latest

# Install Playwright browsers
RUN npx playwright install chromium --with-deps
```

--------------------------------

### Server Debug Logs

Source: https://github.com/executeautomation/mcp-playwright/blob/main/docs/docs/playwright-web/HTTP-SSE-Transport.mdx

Example of the JSON-formatted logs generated by the server for request and session tracking.

```json
{"level":"info","message":"Incoming request","context":{"method":"GET","path":"/mcp"}}
{"level":"info","message":"Transport registered","context":{"sessionId":"abc-123","activeTransports":1}}
{"level":"info","message":"POST message received","context":{"sessionId":"abc-123","endpoint":"/mcp"}}
```

--------------------------------

### Start HTTP/SSE Transport Server

Source: https://github.com/executeautomation/mcp-playwright/blob/main/docs/docs/release.mdx

Launches the mcp-playwright server in standalone HTTP mode with Server-Sent Events support. Ensure the specified port is available.

```bash
playwright-mcp-server --port 8931
```

--------------------------------

### Install Playwright MCP Server with Claude Code

Source: https://github.com/executeautomation/mcp-playwright/blob/main/README.md

Add the Playwright MCP server using Claude's command-line interface with stdio transport.

```bash
claude mcp add --transport stdio playwright npx @executeautomation/playwright-mcp-server
```

--------------------------------

### Execute CRUD Operations Scenario

Source: https://github.com/executeautomation/mcp-playwright/blob/main/docs/docs/playwright-api/Examples.md

A high-level scenario demonstrating POST, GET, PUT, and PATCH operations with token authorization.

```json
// Basic POST request
Perform POST operation for the URL https://api.restful-api.dev/objects with body
{
   "name": "Apple MacBook Pro 16",
   "data": {
      "year": 2024,
      "price": 2499,
      "CPU model": "M4",
      "Hard disk size": "5 TB"
   }
}
And verify if the response has createdAt and id property and store the ID in a variable for future reference say variable productID

// POST request with Bearer token authorization
Perform POST operation for the URL https://api.restful-api.dev/objects with Bearer token "your-token-here" set in the headers
{
        'Content-Type': 'application/json',
        'Authorization': 'Bearer your-token-here'
      },
and body
{
   "name": "Secure MacBook Pro",
   "data": {
      "year": 2024,
      "price": 2999,
      "CPU model": "M4 Pro",
      "Hard disk size": "8 TB",
      "security": "enhanced"
   }
}

Perform GET operation for the created ProductID using URL https://api.restful-api.dev/objects/productID and verify the response has properties like Id, name, data

Perform PUT operation for the created ProductID using URL https://api.restful-api.dev/objects/productID with body {
   "name": "Apple MacBook Pro 16",
   "data": {
      "year": 2025,
      "price": 4099,
      "CPU model": "M5",
      "Hard disk size": "10 TB",
      "color": "Titanium"
   }
}

And verify if the response has createdAt and id property

Perform PATCH operation for the created ProductID using URL https://api.restful-api.dev/objects/productID with body
{
   "name": "Apple MacBook Pro 19 (Limited Edition)"
}
And verify if the response has updatedAt property with value Apple MacBook Pro 19 (Limited Edition)
```

--------------------------------

### Access MCP Inspector URL

Source: https://github.com/executeautomation/mcp-playwright/blob/main/docs/docs/local-setup/Debugging.mdx

Navigate to this local address in your browser after starting the inspector.

```text
http://localhost:5173 🚀
```

--------------------------------

### Analyze page text content

Source: https://github.com/executeautomation/mcp-playwright/blob/main/docs/docs/playwright-web/Examples.md

Example workflow for extracting text to perform content analysis.

```bdd
Given I navigate to website "https://example.com/pricing"
When I extract all visible text from the page
Then I should be able to analyze the text to find pricing information
And I can determine if the "Enterprise" plan mentions "custom pricing"
```

--------------------------------

### Start Code Generation Session

Source: https://github.com/executeautomation/mcp-playwright/blob/main/docs/docs/release.mdx

Initiates a code generation session to record Playwright actions. This tool allows for the conversion of real browser interactions into reusable Playwright tests.

```python
start_codegen_session
```

--------------------------------

### Get Code Generation Session Info

Source: https://context7.com/executeautomation/mcp-playwright/llms.txt

Retrieves information about an active code generation session. Requires a session ID.

```javascript
await get_codegen_session({
  sessionId: "your-session-id"
});
```

--------------------------------

### GET /Playwright_get

Source: https://github.com/executeautomation/mcp-playwright/blob/main/docs/docs/playwright-api/Supported-Tools.mdx

Perform a GET operation on any given API request.

```APIDOC
## GET Playwright_get

### Description
Perform a GET operation on any given API request.

### Parameters
#### Request Body
- **url** (string) - Required - URL to perform the GET operation.
- **token** (string) - Optional - Bearer token for authorization.
- **headers** (object) - Optional - Additional headers to include in the request.

### Response
#### Success Response (200)
- **statusCode** (string) - Status code of the API.
```

--------------------------------

### Execute Login Scenario

Source: https://github.com/executeautomation/mcp-playwright/blob/main/docs/docs/playwright-web/Examples.md

Performs a standard login and navigation flow on a sample application.

```bdd
Given I navigate to website http://eaapp.somee.com and click login link
And I enter username and password as "admin" and "password" respectively and perform login
Then click the Employee List page 
And click "Create New" button and enter realistic employee details to create for Name, Salary, DurationWorked,
Select dropdown for Grade as CLevel and Email.
```

--------------------------------

### Advanced Navigation with Playwright MCP

Source: https://context7.com/executeautomation/mcp-playwright/llms.txt

Navigates to a URL with custom browser type (Firefox), viewport size, timeout, wait condition, and headless mode. Requires specific browser engine setup.

```javascript
await playwright_navigate({
  url: "https://example.com",
  browserType: "firefox",
  width: 1920,
  height: 1080,
  timeout: 60000,
  waitUntil: "networkidle",
  headless: true
});
```

--------------------------------

### Listing and Filtering Devices

Source: https://github.com/executeautomation/mcp-playwright/blob/main/docs/docs/playwright-web/Device-Quick-Reference.md

Commands to list available device presets or filter them via the CLI.

```bash
node scripts/list-devices.cjs
```

```bash
node scripts/list-devices.cjs iphone
node scripts/list-devices.cjs pixel
node scripts/list-devices.cjs ipad
node scripts/list-devices.cjs galaxy
```

--------------------------------

### API GET Request with Bearer Token Authentication

Source: https://github.com/executeautomation/mcp-playwright/blob/main/docs/docs/release.mdx

Perform a GET request to a protected API endpoint using Bearer token authentication. Ensure the token is correctly formatted.

```javascript
await playwright_get({
    url: 'https://api.example.com/protected-data',
    token: 'your-bearer-token'
  });
```

--------------------------------

### Automated Testing

Source: https://github.com/executeautomation/mcp-playwright/blob/main/docs/docs/playwright-web/HTTP-SSE-Transport.mdx

Build the project and execute the provided test script.

```bash
# Clone the repository
git clone https://github.com/executeautomation/mcp-playwright.git
cd mcp-playwright

# Build
npm run build

# Run automated tests
./test-http-mode.sh
```

--------------------------------

### GET /get_codegen_session

Source: https://context7.com/executeautomation/mcp-playwright/llms.txt

Retrieves information about an active code generation session.

```APIDOC
## GET /get_codegen_session

### Description
Get information about an active code generation session.

### Method
GET

### Endpoint
/get_codegen_session

### Request Body
- **sessionId** (string) - Required - The ID of the session to retrieve.
```

--------------------------------

### Docker Deployment

Source: https://github.com/executeautomation/mcp-playwright/blob/main/docs/docs/playwright-web/HTTP-SSE-Transport.mdx

Build and run the server within a Docker container.

```dockerfile
FROM node:20

WORKDIR /app

RUN npm install -g @executeautomation/playwright-mcp-server

EXPOSE 8931

CMD ["playwright-mcp-server", "--port", "8931"]
```

```bash
docker build -t playwright-mcp .
docker run -p 8931:8931 playwright-mcp
```

--------------------------------

### playwright_expect_response

Source: https://context7.com/executeautomation/mcp-playwright/llms.txt

Starts waiting for an HTTP response matching a specific URL pattern.

```APIDOC
## playwright_expect_response

### Description
Start waiting for an HTTP response matching a URL pattern. Use with playwright_assert_response to validate responses.

### Parameters
#### Request Body
- **id** (string) - Required - Unique identifier for the response expectation.
- **url** (string) - Required - URL pattern to match.
```

--------------------------------

### playwright_get

Source: https://context7.com/executeautomation/mcp-playwright/llms.txt

Performs an HTTP GET request with optional authentication and custom headers.

```APIDOC
## playwright_get

### Description
Perform an HTTP GET request with optional authentication and custom headers.

### Parameters
#### Request Body
- **url** (string) - Required - The target URL.
- **token** (string) - Optional - Bearer token for authentication.
- **headers** (object) - Optional - Custom HTTP headers.
```

--------------------------------

### SSE and Unified Endpoints

Source: https://github.com/executeautomation/mcp-playwright/blob/main/docs/docs/playwright-web/HTTP-SSE-Transport.mdx

Connect to the server using legacy SSE or the recommended unified MCP endpoint.

```bash
curl -N http://localhost:8931/sse
```

```bash
curl -N http://localhost:8931/mcp
```

--------------------------------

### Playwright Get Visible Text

Source: https://github.com/executeautomation/mcp-playwright/blob/main/docs/docs/playwright-web/Supported-Tools.mdx

Retrieves the visible text content of the current page.

```APIDOC
## playwright_get_visible_text

### Description
Get the visible text content of the current page.

### Method
GET

### Endpoint
/executeautomation/mcp-playwright/playwright_get_visible_text

### Response
#### Success Response (200)
- **`content`** (string) - The visible text content of the current page, extracted from visible DOM elements. Hidden elements (with display:none or visibility:hidden) are excluded.
```

--------------------------------

### Device Discovery Prompts

Source: https://github.com/executeautomation/mcp-playwright/blob/main/docs/docs/playwright-web/Resize-Prompts-Guide.md

Natural language prompts used to query available device configurations.

```text
What iPhone models can I test on?
```

```text
Show me all available Android devices
```

```text
What desktop browser options do I have?
```

--------------------------------

### Automate Multiple Browser Engines

Source: https://github.com/executeautomation/mcp-playwright/blob/main/docs/docs/playwright-web/Examples.md

Demonstrates switching between Firefox and WebKit engines to compare rendering.

```bdd
Given I navigate to website "https://example.com" using the "firefox" browser
And I take a screenshot named "firefox-example"
Then I navigate to website "https://example.com" using the "webkit" browser 
And I take a screenshot named "webkit-example"
```

--------------------------------

### Command-Line Interface Usage

Source: https://github.com/executeautomation/mcp-playwright/blob/main/docs/docs/playwright-web/HTTP-SSE-Transport.mdx

Run the server in stdio or HTTP mode using various CLI flags.

```bash
playwright-mcp-server [OPTIONS]

OPTIONS:
  --port <number>    Run in HTTP mode on the specified port
  --help, -h         Show help message

EXAMPLES:
  # Run in stdio mode (default)
  playwright-mcp-server

  # Run in HTTP mode on port 8931
  playwright-mcp-server --port 8931

  # Run in HTTP mode on custom port
  playwright-mcp-server --port 3000
```

--------------------------------

### Clone Repository

Source: https://github.com/executeautomation/mcp-playwright/blob/main/docs/docs/local-setup/Installation.mdx

Download the source code from the official repository.

```bash
git clone https://github.com/executeautomation/mcp-playwright.git
```

--------------------------------

### Manual Dimension Prompts

Source: https://github.com/executeautomation/mcp-playwright/blob/main/docs/docs/playwright-web/Resize-Prompts-Guide.md

Prompts for setting specific viewport dimensions.

```text
Resize to 1024x768
```

```text
Set viewport to 800x600
```

```text
Test at 1366x768 resolution
```

```text
Test at small tablet size: 768x1024
```

```text
Show me the mobile breakpoint: 375x667
```

```text
Resize to standard HD: 1920x1080
```

--------------------------------

### Complete Login Test Workflow

Source: https://context7.com/executeautomation/mcp-playwright/llms.txt

Demonstrates a full login test automation workflow, including navigation, filling credentials, API response assertion, and cleanup.

```javascript
// 1. Navigate to the login page
await playwright_navigate({
  url: "https://myapp.example.com/login",
  browserType: "chromium",
  headless: false
});
```

```javascript
// 2. Take initial screenshot
await playwright_screenshot({ name: "login-page" });
```

```javascript
// 3. Fill in credentials
await playwright_fill({
  selector: "#email",
  value: "user@example.com"
});
```

```javascript
await playwright_fill({
  selector: "#password",
  value: "password123"
});
```

```javascript
// 4. Set up response expectation before clicking
await playwright_expect_response({
  id: "auth-response",
  url: "**/api/auth"
});
```

```javascript
// 5. Click login button
await playwright_click({ selector: "#login-submit" });
```

```javascript
// 6. Assert the authentication response
await playwright_assert_response({
  id: "auth-response",
  value: "token"
});
```

```javascript
// 7. Take screenshot of dashboard
await playwright_screenshot({
  name: "dashboard",
  fullPage: true
});
```

```javascript
// 8. Get page content for verification
await playwright_get_visible_text({});
```

```javascript
// 9. Close browser
await playwright_close({});
```

--------------------------------

### Playwright Get Visible HTML

Source: https://github.com/executeautomation/mcp-playwright/blob/main/docs/docs/playwright-web/Supported-Tools.mdx

Retrieves the HTML content of the current page, with options for filtering and cleaning.

```APIDOC
## playwright_get_visible_html

### Description
Get the HTML content of the current page.

### Method
POST

### Endpoint
/executeautomation/mcp-playwright/playwright_get_visible_html

### Parameters
#### Request Body
- **`selector`** (string) - Optional - CSS selector to limit the HTML to a specific container. If provided, only returns the HTML for that element.
- **`removeScripts`** (boolean) - Optional, default: false - Remove all script tags from the HTML to reduce noise.
- **`removeComments`** (boolean) - Optional, default: false - Remove all HTML comments to clean up the output.
- **`removeStyles`** (boolean) - Optional, default: false - Remove all style tags from the HTML.
- **`removeMeta`** (boolean) - Optional, default: false - Remove all meta tags from the HTML head section.
- **`minify`** (boolean) - Optional, default: false - Minify the HTML output by removing extra whitespace.
- **`cleanHtml`** (boolean) - Optional, default: false - Convenience option that combines removeScripts, removeComments, removeStyles, and removeMeta for a cleaner HTML output.

### Response
#### Success Response (200)
- **`content`** (string) - The HTML content of the current page, optionally filtered based on the provided parameters.
```

--------------------------------

### Perform Bearer Token Authentication

Source: https://github.com/executeautomation/mcp-playwright/blob/main/docs/docs/playwright-api/Examples.md

Use the token parameter to authenticate GET, POST, and DELETE requests.

```typescript
// GET request with Bearer token
await playwright_get({
  url: 'https://api.example.com/protected-data',
  token: 'your-bearer-token-here'
});

// POST request with Bearer token
await playwright_post({
  url: 'https://api.example.com/create',
  value: '{"name":"test"}',
  token: 'your-bearer-token-here'
});

// DELETE request with Bearer token
await playwright_delete({
  url: 'https://api.example.com/resource/123',
  token: 'your-bearer-token-here'
});
```

--------------------------------

### Build Docker Image with Specific Tag

Source: https://github.com/executeautomation/mcp-playwright/blob/main/DOCKER.md

Build the Docker image and tag it with a specific version number.

```bash
docker build -t mcp-playwright:1.0.6 .
```

--------------------------------

### Get Visible HTML Content

Source: https://context7.com/executeautomation/mcp-playwright/llms.txt

Retrieve the page HTML with options to clean, minify, or filter specific elements.

```javascript
// Get HTML with scripts removed (default)
await playwright_get_visible_html({});

// Get HTML for specific element
await playwright_get_visible_html({
  selector: "#main-content"
});

// Get cleaned and minified HTML
await playwright_get_visible_html({
  cleanHtml: true,
  minify: true,
  removeComments: true,
  removeStyles: true,
  removeMeta: true
});

// Keep scripts in output
await playwright_get_visible_html({
  removeScripts: false,
  maxLength: 50000
});
```

--------------------------------

### Create Custom Docker Network

Source: https://github.com/executeautomation/mcp-playwright/blob/main/DOCKER.md

Create a custom Docker network and run the Playwright MCP server container on it for custom network configurations.

```bash
docker network create mcp-network
docker run -i --rm --network mcp-network mcp-playwright:latest
```

--------------------------------

### Get Visible Page Text

Source: https://context7.com/executeautomation/mcp-playwright/llms.txt

Extract visible text content from the current page, excluding hidden elements.

```javascript
// Get all visible text
await playwright_get_visible_text({});

// Get text with custom max length
await playwright_get_visible_text({ maxLength: 5000 });
```

--------------------------------

### Health Monitoring with jq

Source: https://github.com/executeautomation/mcp-playwright/blob/main/docs/docs/playwright-web/HTTP-SSE-Transport.mdx

Format health status output using jq.

```bash
curl http://localhost:8931/health | jq
```

--------------------------------

### Responsive Web Design Testing

Source: https://github.com/executeautomation/mcp-playwright/blob/main/docs/docs/playwright-web/Resize-Prompts-Guide.md

Demonstrates navigating to a URL and resizing the viewport to specific device dimensions.

```javascript
await playwright_navigate({ url: "your-site.com" });
await playwright_resize({ device: "iPhone 13" });
// AI tests menu functionality
```

```javascript
await playwright_resize({ device: "iPad Pro 11" });
```

--------------------------------

### Remote Server Deployment

Source: https://github.com/executeautomation/mcp-playwright/blob/main/docs/docs/playwright-web/HTTP-SSE-Transport.mdx

Deploy the server on a remote machine and configure the client connection.

```bash
# On remote server
ssh user@remote-server
playwright-mcp-server --port 8931

# Configure client to connect remotely
{
  "mcpServers": {
    "playwright": {
      "url": "http://remote-server:8931/mcp"
    }
  }
}
```

--------------------------------

### Run Server with Environment Variables

Source: https://github.com/executeautomation/mcp-playwright/blob/main/DOCKER.md

Pass environment variables to the Docker container to configure the server, such as skipping browser downloads.

```bash
docker run -i --rm \
  -e PLAYWRIGHT_SKIP_BROWSER_DOWNLOAD=1 \
  mcp-playwright:latest
```

--------------------------------

### Configure Claude Desktop

Source: https://github.com/executeautomation/mcp-playwright/blob/main/docs/docs/local-setup/Installation.mdx

Update the Claude Desktop configuration file to point to the local MCP server directory.

```json
{
  "mcpServers": {
    "playwright": {
      "command": "npx",
      "args": [
        "--directory",
        "/your-playwright-mcp-server-clone-directory",
        "run",
        "@executeautomation/playwright-mcp-server"
      ]
    }
  }
}
```

--------------------------------

### Background Service Management

Source: https://github.com/executeautomation/mcp-playwright/blob/main/docs/docs/playwright-web/HTTP-SSE-Transport.mdx

Run the server as a persistent background process and monitor logs.

```bash
# Using nohup
nohup playwright-mcp-server --port 8931 > server.log 2>&1 &

# Check status
curl http://localhost:8931/health

# View logs
tail -f server.log
```

--------------------------------

### Get Visible HTML Content

Source: https://github.com/executeautomation/mcp-playwright/blob/main/docs/docs/release.mdx

Retrieves the complete HTML content of the current web page. By default, script tags are removed and the output is truncated.

```python
playwright_get_visible_html
```

--------------------------------

### Cross-Device Comparison Prompt

Source: https://github.com/executeautomation/mcp-playwright/blob/main/docs/docs/playwright-web/Resize-Prompts-Guide.md

Prompt for comparing layout across multiple devices.

```text
Compare how the navigation menu looks on:
- iPhone 13
- iPad Pro 11  
- Desktop Chrome

Take screenshots and tell me if there are any layout issues.
```

--------------------------------

### Context-Aware Testing Prompt

Source: https://github.com/executeautomation/mcp-playwright/blob/main/docs/docs/playwright-web/Resize-Prompts-Guide.md

Prompt for switching devices during an active test session.

```text
Can you switch to iPhone SE (smaller screen) and check if the button is still clickable?
```

--------------------------------

### Quick Device Switch Prompts

Source: https://github.com/executeautomation/mcp-playwright/blob/main/docs/docs/playwright-web/Resize-Prompts-Guide.md

Short prompts for switching devices.

```text
Switch to Pixel 7
```

```text
Try this on Galaxy S24 instead
```

```text
Show me desktop view
```

--------------------------------

### Playwright MCP Server API Testing Support

Source: https://github.com/executeautomation/mcp-playwright/blob/main/docs/docs/release.mdx

The Playwright MCP Server now includes support for various API testing methods, including GET, POST, PUT, PATCH, and DELETE requests.

```plaintext
GET request
POST request
PUT request
PATCH request
DELETE request
```

--------------------------------

### Desktop Browser Testing Prompts

Source: https://github.com/executeautomation/mcp-playwright/blob/main/docs/docs/playwright-web/Resize-Prompts-Guide.md

Natural language prompts for testing desktop browser viewports.

```text
Test in Desktop Chrome
```

```text
Switch to Desktop Firefox
```

```text
Show desktop Safari view
```

```text
Test at full desktop resolution
```

--------------------------------

### Generate Playwright Test Spec from Recorded Actions

Source: https://github.com/executeautomation/mcp-playwright/blob/main/docs/docs/playwright-web/Recording-Actions.mdx

This TypeScript code demonstrates a generated Playwright test spec. It navigates to a website, fills in login credentials, clicks the login button, and verifies successful login by checking the URL. Ensure Playwright is installed and configured.

```typescript
import { test, expect } from '@playwright/test';

test('LoginTest_2024-03-23', async ({ page }) => {
  // Navigate to the login page
  await page.goto('https://www.saucedemo.com');

  // Fill in username
  await page.fill('#user-name', 'standard_user');

  // Fill in password
  await page.fill('#password', 'secret_sauce');

  // Click login button
  await page.click('#login-button');

  // Verify successful login
  await expect(page).toHaveURL(/.*inventory.html/);
});
```

--------------------------------

### Device Recommendation Prompt

Source: https://github.com/executeautomation/mcp-playwright/blob/main/docs/docs/playwright-web/Resize-Prompts-Guide.md

Prompt to request expert advice on which devices to include in a test suite.

```text
Which device should I test for mobile users?
```

--------------------------------

### Run Server with User Permissions for Mounted Volumes

Source: https://github.com/executeautomation/mcp-playwright/blob/main/DOCKER.md

Run the Docker container with specific user and group IDs to resolve permission issues when using mounted volumes.

```bash
docker run -i --rm \
  -v $(pwd)/data:/app/data \
  --user $(id -u):$(id -g) \
  mcp-playwright:latest
```

--------------------------------

### Run Evaluations

Source: https://github.com/executeautomation/mcp-playwright/blob/main/README.md

Execute evaluations by providing the necessary API key as an environment variable.

```bash
OPENAI_API_KEY=your-key  npx mcp-eval src/evals/evals.ts src/tools/codegen/index.ts
```

--------------------------------

### Execute Test Suites

Source: https://github.com/executeautomation/mcp-playwright/blob/main/README.md

Commands for running tests with or without coverage reports.

```bash
# Run tests using the custom script (with coverage)
node run-tests.cjs

# Run tests using npm scripts
npm test           # Run tests without coverage
npm run test:coverage  # Run tests with coverage
npm run test:custom    # Run tests with custom script (same as node run-tests.cjs)
```

--------------------------------

### Run Server with Resource Limits

Source: https://github.com/executeautomation/mcp-playwright/blob/main/DOCKER.md

Limit the CPU and memory resources allocated to the Docker container to manage resource consumption.

```bash
docker run -i --rm \
  --cpus="2.0" \
  --memory="2g" \
  mcp-playwright:latest
```

--------------------------------

### Combine Token and Headers

Source: https://github.com/executeautomation/mcp-playwright/blob/main/docs/docs/playwright-api/Examples.md

Demonstrates how to combine token and headers, noting that custom Authorization headers override the token parameter.

```typescript
// Token takes precedence if no Authorization header in custom headers
await playwright_get({
  url: 'https://api.example.com/data',
  token: 'bearer-token',
  headers: {
    'X-Custom-Header': 'value'
  }
});

// Custom Authorization header overrides token parameter
await playwright_get({
  url: 'https://api.example.com/data',
  token: 'this-will-be-ignored',
  headers: {
    'Authorization': 'Basic xyz123',  // This takes precedence
    'X-Custom-Header': 'value'
  }
});
```

--------------------------------

### Manage Server Ports

Source: https://github.com/executeautomation/mcp-playwright/blob/main/docs/docs/playwright-web/HTTP-SSE-Transport.mdx

Commands to identify port conflicts and switch to an alternative port.

```bash
lsof -i :8931

playwright-mcp-server --port 9000
```

--------------------------------

### Run Server with Volume Mounts

Source: https://github.com/executeautomation/mcp-playwright/blob/main/DOCKER.md

Mount local directories into the Docker container to persist data or share files, such as for logs or screenshots.

```bash
docker run -i --rm \
  -v $(pwd)/data:/app/data \
  mcp-playwright:latest
```

--------------------------------

### Build MCP Playwright from Source in Docker

Source: https://github.com/executeautomation/mcp-playwright/blob/main/DOCKER.md

This Dockerfile demonstrates how to build the MCP Playwright project from source within a Docker container. It uses a multi-stage build to keep the final image lean, copying only necessary artifacts from the builder stage.

```dockerfile
FROM node:20-alpine AS builder

WORKDIR /app
COPY package*.json ./
ENV PLAYWRIGHT_SKIP_BROWSER_DOWNLOAD=1
RUN npm ci

COPY src ./src
COPY tsconfig.json ./
RUN npm run build

FROM node:20-alpine
WORKDIR /app
COPY --from=builder /app/dist ./dist
COPY --from=builder /app/package*.json ./
RUN npm ci --only=production
CMD ["node", "dist/index.js"]
```

--------------------------------

### Debug SSE Connections

Source: https://github.com/executeautomation/mcp-playwright/blob/main/docs/docs/playwright-web/HTTP-SSE-Transport.mdx

Verify that the Server-Sent Events stream is established and sessions are registered.

```bash
curl -N http://localhost:8931/mcp
```

```bash
grep "Transport registered" server.log
```

--------------------------------

### Concurrent Client Testing

Source: https://github.com/executeautomation/mcp-playwright/blob/main/docs/docs/playwright-web/HTTP-SSE-Transport.mdx

Verify that the server tracks multiple active sessions in HTTP mode.

```bash
# Start server
playwright-mcp-server --port 8931

# Client 1 connects
# Client 2 connects
# Client 3 connects

# Check active sessions
curl http://localhost:8931/health
# Returns: {"status":"ok","activeSessions":3}
```

--------------------------------

### Monitor Server Metrics

Source: https://github.com/executeautomation/mcp-playwright/blob/main/docs/docs/playwright-web/HTTP-SSE-Transport.mdx

Retrieve memory usage and active session counts via the metrics and health endpoints.

```bash
curl http://localhost:8932/metrics | jq '.memory'

curl http://localhost:8931/health | jq '.activeSessions'
```

--------------------------------

### Tablet Device Testing Prompts

Source: https://github.com/executeautomation/mcp-playwright/blob/main/docs/docs/playwright-web/Resize-Prompts-Guide.md

Natural language prompts for testing tablet device viewports.

```text
Test this on iPad Pro 11
```

```text
Show me the tablet view using iPad Air
```

```text
Switch to iPad Pro 12.9 in landscape
```

```text
Test on Galaxy Tab S4
```

```text
Show tablet view on Pixel Tablet
```

--------------------------------

### Take Screenshot with Custom Directory

Source: https://context7.com/executeautomation/mcp-playwright/llms.txt

Takes a full-page screenshot and saves it to a custom directory specified by `downloadsDir`. Useful for organizing screenshots.

```javascript
// Screenshot with custom directory
await playwright_screenshot({
  name: "report",
  fullPage: true,
  downloadsDir: "/path/to/screenshots"
});
```

--------------------------------

### Orientation Testing Prompt

Source: https://github.com/executeautomation/mcp-playwright/blob/main/docs/docs/playwright-web/Resize-Prompts-Guide.md

Prompt for testing portrait and landscape modes.

```text
Test the page in both portrait and landscape on iPhone 14:
1. Portrait mode first
2. Then landscape mode
Check if content reflows properly.
```

--------------------------------

### Monitoring Metrics

Source: https://github.com/executeautomation/mcp-playwright/blob/main/docs/docs/playwright-web/HTTP-SSE-Transport.mdx

Access health and metrics data, typically hosted on the port following the main server port.

```bash
curl http://localhost:8932/health
curl http://localhost:8932/metrics
```

--------------------------------

### Android Phone Testing Prompts

Source: https://github.com/executeautomation/mcp-playwright/blob/main/docs/docs/playwright-web/Resize-Prompts-Guide.md

Natural language prompts for testing Android device viewports.

```text
Test on Pixel 7
```

```text
Show me this on Galaxy S24
```

```text
Switch to Pixel 5 view
```

```text
Test on Galaxy S9+ in landscape
```

--------------------------------

### Run Server in Interactive Mode

Source: https://github.com/executeautomation/mcp-playwright/blob/main/DOCKER.md

Run the Docker container interactively, which is recommended for MCP servers that communicate via stdin/stdout. The container is automatically removed upon exit.

```bash
docker run -i --rm mcp-playwright:latest
```

--------------------------------

### Navigate Forward in Browser

Source: https://github.com/executeautomation/mcp-playwright/blob/main/docs/docs/release.mdx

Performs a forward navigation in the browser's history. This tool complements the backward navigation functionality.

```python
playwright_go_forward
```

--------------------------------

### Simulate Keyboard Input

Source: https://github.com/executeautomation/mcp-playwright/blob/main/docs/docs/playwright-web/Examples.md

Focuses on an input field and triggers a key press event.

```bdd
Given I navigate to website "https://example.com/form"
When I focus on the input field with id "search-box"
And I press the "Enter" key
Then the search results should appear
```

--------------------------------

### Take Basic Screenshot

Source: https://context7.com/executeautomation/mcp-playwright/llms.txt

Takes a screenshot of the current page and saves it with the specified name. By default, it saves to the Downloads folder.

```javascript
// Basic screenshot
await playwright_screenshot({ name: "homepage" });
```

--------------------------------

### Emulate Devices with Playwright MCP

Source: https://github.com/executeautomation/mcp-playwright/blob/main/README.md

Use the playwright_resize function to emulate various devices. Supports automatic user-agent, touch support, and device pixel ratio. You can also specify orientation.

```javascript
await playwright_resize({ device: "iPhone 13" });
```

```javascript
await playwright_resize({ device: "iPad Pro 11", orientation: "landscape" });
```

```javascript
await playwright_resize({ device: "Desktop Chrome" });
```

--------------------------------

### Log Analysis

Source: https://github.com/executeautomation/mcp-playwright/blob/main/docs/docs/playwright-web/HTTP-SSE-Transport.mdx

Filter and monitor server logs in real-time.

```bash
# View logs in real-time
tail -f server.log

# Filter by level
grep '"level":"error"' server.log

# Track specific session
grep '"sessionId":"abc-123"' server.log
```

--------------------------------

### Cross-Browser Testing

Source: https://github.com/executeautomation/mcp-playwright/blob/main/docs/docs/playwright-web/Resize-Prompts-Guide.md

Automates screenshot capture across different desktop browser environments.

```javascript
await playwright_resize({ device: "Desktop Chrome" });
await playwright_screenshot({ name: "chrome-view" });
await playwright_resize({ device: "Desktop Firefox" });
await playwright_screenshot({ name: "firefox-view" });
await playwright_resize({ device: "Desktop Safari" });
await playwright_screenshot({ name: "safari-view" });
```

--------------------------------

### Mobile Device Testing Prompts

Source: https://github.com/executeautomation/mcp-playwright/blob/main/docs/docs/playwright-web/Resize-Prompts-Guide.md

Natural language prompts for testing specific mobile device viewports.

```text
Test this page on iPhone 13
```

```text
Switch to iPhone 14 view
```

```text
Show me how this looks on iPhone SE
```

```text
Test on iPhone 13 in landscape mode
```

```text
Rotate to landscape orientation on iPhone 15
```

```text
Switch iPhone to portrait mode
```

--------------------------------

### SSE Stream (Legacy)

Source: https://github.com/executeautomation/mcp-playwright/blob/main/docs/docs/playwright-web/HTTP-SSE-Transport.mdx

Establishes a Server-Sent Events (SSE) connection for streaming data. This is a legacy endpoint.

```APIDOC
## GET /sse (Legacy)

### Description
Establishes an SSE connection. This endpoint is intended for legacy use cases and may be deprecated in favor of the unified endpoint.

### Method
GET

### Endpoint
/sse

### Notes
Use the `-N` flag with curl to disable output buffering for real-time streaming.
```

--------------------------------

### Claude Desktop MCP Server Configuration

Source: https://github.com/executeautomation/mcp-playwright/blob/main/DOCKER.md

Configure Claude Desktop to use the Dockerized Playwright MCP server by specifying the Docker command and arguments in the configuration file.

```json
{
  "mcpServers": {
    "playwright-docker": {
      "command": "docker",
      "args": ["run", "-i", "--rm", "mcp-playwright:latest"]
    }
  }
}
```

--------------------------------

### Manual Testing and Inspection

Source: https://github.com/executeautomation/mcp-playwright/blob/main/docs/docs/playwright-web/HTTP-SSE-Transport.mdx

Manually verify endpoints or use the official MCP inspector.

```bash
# Terminal 1: Start server
playwright-mcp-server --port 8931

# Terminal 2: Test endpoints
curl http://localhost:8931/health
curl -N http://localhost:8931/mcp | head -5
curl http://localhost:8932/metrics
```

```bash
# Start server
playwright-mcp-server --port 8931

# Test with official MCP inspector
npx @modelcontextprotocol/inspector http://localhost:8931/mcp
```

--------------------------------

### Resolve npx Output Issues

Source: https://github.com/executeautomation/mcp-playwright/blob/main/docs/docs/playwright-web/HTTP-SSE-Transport.mdx

Commands to fix missing console output or verify server status when using npx.

```bash
npx @executeautomation/playwright-mcp-server@latest --port 8931
```

```bash
npm install -g @executeautomation/playwright-mcp-server
playwright-mcp-server --port 8931
```

```bash
curl http://localhost:8931/health
```

--------------------------------

### Verify Transport Sequence

Source: https://github.com/executeautomation/mcp-playwright/blob/main/docs/docs/playwright-web/HTTP-SSE-Transport.mdx

Log sequence verification for troubleshooting 'No transport found' errors.

```bash
# Check logs for "Transport registered" - should appear before POST messages
# Look for this sequence:
# 1. "Incoming request" GET /mcp
# 2. "SSE connection request received"
# 3. "Transport registered" with sessionId
# 4. "POST message received" with same sessionId

# If sequence is wrong, client needs to reconnect
```

--------------------------------

### Define Cursor Rules for Automated Testing

Source: https://github.com/executeautomation/mcp-playwright/blob/main/docs/docs/testing-videos/AIAgents.mdx

Configures AI agents to perform specific UI navigation, data creation, and database verification tasks when triggered by the /test command.

```js
When I say /test, then perform following

Navigate to http://localhost:8000/Product/List. 

Create product by clicking Create link . 

Then create a product with some realistic data for Name, Price and Select ProductType as CPU and click create input type with id as Create. 

Check the Database for the created record

Use the Schema: http://localhost:8001/swagger/v1/swagger.json

Here is the baseURL of API Base URL: http://localhost:8001/

Also check the API which performs the GET operation if the values are correctly retreived for the created product from the above schema
```

--------------------------------

### Dynamic Metrics Collection

Source: https://github.com/executeautomation/mcp-playwright/blob/main/docs/docs/playwright-web/HTTP-SSE-Transport.mdx

Access metrics using the dynamically assigned monitoring port.

```bash
# Server output shows:
# Monitoring HTTP server listening on port 59055
# Health check: http://localhost:59055/health
# Metrics: http://localhost:59055/metrics
```

```bash
# Use the port from server output
curl http://localhost:59055/metrics | jq
```

--------------------------------

### Navigation with WebKit Browser

Source: https://context7.com/executeautomation/mcp-playwright/llms.txt

Navigates to a URL using the WebKit browser engine, with the option to run in headed mode. Useful for testing cross-browser compatibility.

```javascript
await playwright_navigate({
  url: "https://example.com",
  browserType: "webkit",
  headless: false
});
```

--------------------------------

### Bug Investigation Workflow

Source: https://github.com/executeautomation/mcp-playwright/blob/main/docs/docs/playwright-web/Resize-Prompts-Guide.md

Captures screenshots across multiple mobile devices to reproduce layout issues.

```javascript
await playwright_navigate({ url: "your-site.com/checkout" });
await playwright_resize({ device: "iPhone 13" });
await playwright_screenshot({ name: "checkout-iphone13" });
await playwright_resize({ device: "iPhone SE" });
await playwright_screenshot({ name: "checkout-iphoneSE" });
```

--------------------------------

### Save Page as PDF

Source: https://github.com/executeautomation/mcp-playwright/blob/main/docs/docs/playwright-web/Examples.md

Exports the current page to a PDF file, with options for custom formatting and margins.

```bdd
Given I navigate to website "https://example.com/report"
When I save the current page as a PDF in "/downloads" folder with name "report.pdf"
Then I should see confirmation that the PDF was saved
```

```bdd
Given I navigate to website "https://example.com/invoice"
When I save the page as PDF with the following settings:
  | Setting          | Value     |
  | ----------------- | --------- |
  | Output Path      | /downloads |
  | Filename         | invoice.pdf |
  | Format           | Letter    |
  | Print Background | true      |
  | Top Margin       | 2cm       |
  | Right Margin     | 1cm       |
  | Bottom Margin    | 2cm       |
  | Left Margin      | 1cm       |
Then I should see confirmation that the PDF was saved
```

--------------------------------

### Press Keyboard Key

Source: https://github.com/executeautomation/mcp-playwright/blob/main/docs/docs/release.mdx

Simulates pressing a keyboard key. This can be used with or without focusing on a specific element.

```python
playwright_press_key
```

--------------------------------

### Run as Non-Root User in Dockerfile

Source: https://github.com/executeautomation/mcp-playwright/blob/main/DOCKER.md

Add this to your Dockerfile to run the container as a non-root user, enhancing security. Ensure the 'node' user has the necessary permissions.

```dockerfile
FROM mcp-playwright:latest
USER node
```

--------------------------------

### Deploy via SSH Tunneling

Source: https://github.com/executeautomation/mcp-playwright/blob/main/docs/docs/playwright-web/HTTP-SSE-Transport.mdx

Securely expose the server running on a remote machine to a local environment.

```bash
playwright-mcp-server --port 8931

ssh -L 8931:localhost:8931 user@remote-server
```

```bash
ssh -L 8931:localhost:8931 user@remote-server
```