n8n-nodes-playwright-test-executor

This is an n8n community node for AI-powered browser test automation using Playwright. It lets you execute AI-generated test specifications in your n8n workflows.

n8n is a fair-code licensed workflow automation platform.

Installation Operations Use Cases Compatibility Resources

Installation

Follow the installation guide in the n8n community nodes documentation.

Install via n8n UI:

1. Go to Settings → Community Nodes
2. Click Install
3. Enter: n8n-nodes-playwright-test-executor

Or install via command line:

npm install n8n-nodes-playwright-test-executor

Note: By default the package runs browsers locally and will attempt to download browser binaries (~1GB) on first use. If you are running n8n in a container and don't want to install browsers inside it, use Remote Execution Mode instead (see below).

Operations

This node supports the following operations:

Standard Operations

• Navigate: Go to a specified URL
• Take Screenshot: Capture a screenshot of a webpage
• Get Text: Extract text from an element using CSS selector
• Click Element: Click on an element using CSS selector
• Fill Form: Fill a form field using CSS selector

AI-Powered Operation

• Execute Test Specification: Execute AI-generated test cases in JSON format

Use Cases

AI-Powered Test Automation

This node is designed to work with AI models (like Gemini, GPT-4, Claude) to create automated testing workflows:

HTML Content → AI Model → Test Specification → Playwright Execution → Results

Example Workflow:

1. HTTP Request node: Fetch HTML from target website
2. AI Agent node (Gemini/OpenAI): Analyze HTML and generate test cases
3. Playwright node: Execute test specification
4. Code node: Format results
5. Slack/Email node: Send test report

Test Specification Format

The "Execute Test Specification" operation accepts JSON with the following structure:

[
  {
    "test_name": "Verify homepage loads correctly",
    "steps": [
      {
        "action": "load_page",
        "description": "Navigate to homepage"
      },
      {
        "action": "assert_element_present",
        "selector": "h1",
        "description": "Check main heading exists"
      },
      {
        "action": "assert_text_present",
        "text": "Welcome",
        "description": "Verify welcome message"
      }
    ],
    "expected_result": "Homepage loads with heading and welcome message"
  }
]

Supported Actions

Navigation Actions

| Action | Parameters | Description | Example | |--------|-----------|-------------|---------| | load_page | - | Navigate to base URL | {"action": "load_page"} | | go_back | - | Navigate back in history | {"action": "go_back"} | | go_forward | - | Navigate forward in history | {"action": "go_forward"} | | reload_page | - | Reload current page | {"action": "reload_page"} |

Element Interaction Actions

| Action | Parameters | Description | Example | |--------|-----------|-------------|---------| | click | selector or link | Click element or link text | {"action": "click", "selector": "#button"} | | fill_input | selector, value | Type text into input field | {"action": "fill_input", "selector": "#email", "value": "[email protected]"} | | select_option | selector, value | Select dropdown option | {"action": "select_option", "selector": "#country", "value": "USA"} | | check_checkbox | selector | Check a checkbox | {"action": "check_checkbox", "selector": "#agree"} | | uncheck_checkbox | selector | Uncheck a checkbox | {"action": "uncheck_checkbox", "selector": "#agree"} | | hover | selector | Hover mouse over element | {"action": "hover", "selector": ".menu"} | | press_key | key | Press keyboard key | {"action": "press_key", "key": "Enter"} | | submit_form | - | Submit a form | {"action": "submit_form"} | | scroll_to_element | selector | Scroll element into view | {"action": "scroll_to_element", "selector": "#footer"} |

Assertion Actions

| Action | Parameters | Description | Example | |--------|-----------|-------------|---------| | assert_element_present | selector | Verify element exists | {"action": "assert_element_present", "selector": "h1"} | | assert_element_not_present | selector | Verify element does not exist | {"action": "assert_element_not_present", "selector": ".error"} | | assert_element_visible | selector | Verify element is visible | {"action": "assert_element_visible", "selector": ".modal"} | | assert_text_present | text | Verify text exists on page | {"action": "assert_text_present", "text": "Welcome"} | | assert_url_contains | text | Verify URL contains text | {"action": "assert_url_contains", "text": "/dashboard"} | | assert_page_title | text | Verify page title contains text | {"action": "assert_page_title", "text": "Dashboard"} | | assert_element_count | selector, count | Verify number of matching elements | {"action": "assert_element_count", "selector": ".item", "count": 5} | | assert_attribute | selector, attribute, value | Verify element attribute value | {"action": "assert_attribute", "selector": "#link", "attribute": "href", "value": "/home"} |

Wait Actions

| Action | Parameters | Description | Example | |--------|-----------|-------------|---------| | wait_for_element | selector, timeout (optional) | Wait for element to appear | {"action": "wait_for_element", "selector": ".loading", "timeout": 5000} |

Utility Actions

| Action | Parameters | Description | Example | |--------|-----------|-------------|---------| | take_screenshot | - | Take screenshot (returns base64) | {"action": "take_screenshot"} | | upload_file | selector, file_path | Upload file to input | {"action": "upload_file", "selector": "#upload", "file_path": "/path/file.pdf"} | | clear_input | selector | Clear text from input field | {"action": "clear_input", "selector": "#search"} | | wait | timeout | Wait for specified milliseconds | {"action": "wait", "timeout": 2000} | | execute_script | script | Execute JavaScript code | {"action": "execute_script", "script": "return document.title"} | | set_viewport | width, height | Set browser viewport size | {"action": "set_viewport", "width": 1920, "height": 1080} |

Advanced Interaction Actions

| Action | Parameters | Description | Example | |--------|-----------|-------------|---------| | double_click | selector | Double-click element | {"action": "double_click", "selector": ".item"} | | right_click | selector | Right-click element | {"action": "right_click", "selector": ".menu-item"} | | drag_and_drop | source_selector, target_selector | Drag element to target | {"action": "drag_and_drop", "source_selector": "#item1", "target_selector": "#drop-zone"} | | focus | selector | Focus on element | {"action": "focus", "selector": "#input"} | | blur | selector | Remove focus from element | {"action": "blur", "selector": "#input"} |

Dialog Actions

| Action | Parameters | Description | Example | |--------|-----------|-------------|---------| | accept_dialog | - | Accept next browser alert/confirm | {"action": "accept_dialog"} | | dismiss_dialog | - | Dismiss next browser alert/confirm | {"action": "dismiss_dialog"} |

Frame Actions

| Action | Parameters | Description | Example | |--------|-----------|-------------|---------| | switch_to_iframe | selector | Switch context to iframe | {"action": "switch_to_iframe", "selector": "#my-iframe"} | | switch_to_main | - | Switch back to main frame | {"action": "switch_to_main"} |

Cookie Actions

| Action | Parameters | Description | Example | |--------|-----------|-------------|---------| | set_cookie | name, value, domain, path | Set browser cookie | {"action": "set_cookie", "name": "session", "value": "abc123"} | | get_cookie | name | Get cookie value | {"action": "get_cookie", "name": "session"} | | clear_cookies | - | Clear all cookies | {"action": "clear_cookies"} |

Extended Assertions

| Action | Parameters | Description | Example | |--------|-----------|-------------|---------| | assert_element_enabled | selector | Verify element is enabled | {"action": "assert_element_enabled", "selector": "#submit"} | | assert_element_disabled | selector | Verify element is disabled | {"action": "assert_element_disabled", "selector": "#submit"} | | assert_checkbox_checked | selector | Verify checkbox is checked | {"action": "assert_checkbox_checked", "selector": "#agree"} | | assert_checkbox_unchecked | selector | Verify checkbox is unchecked | {"action": "assert_checkbox_unchecked", "selector": "#agree"} |

Data Extraction Actions

| Action | Parameters | Description | Example | |--------|-----------|-------------|---------| | get_text | selector | Extract text from element | {"action": "get_text", "selector": ".result"} | | get_attribute | selector, attribute | Extract element attribute | {"action": "get_attribute", "selector": "#link", "attribute": "href"} |

Browser Options

• Choose between Chromium, Firefox, or WebKit
• Configure headless mode (default: true)
• Adjust operation speed with slow motion option

Execution Mode

| Mode | Description | |------|-------------| | Local | Launches a browser process on the same machine as n8n. Requires browser binaries to be installed (~1GB). | | Remote | Connects to a Playwright server running in a separate container or machine via WebSocket. No browser binaries needed on the n8n host. |

Remote Execution Setup

Run a Playwright server in a separate container:

# docker-compose.yml
services:
  n8n:
    image: n8nio/n8n
    # ...

  playwright-server:
    image: mcr.microsoft.com/playwright:v1.49.0-noble
    command: npx playwright run-server --port 3000 --host 0.0.0.0
    ports:
      - "3000:3000"

Then in the n8n node:

1. Set Execution Mode to Remote
2. Set Remote WebSocket URL to ws://playwright-server:3000

This is the recommended approach when running n8n in Docker, as it keeps browser dependencies isolated in a dedicated container.

Results Format

The node returns detailed execution results:

{
  "summary": {
    "total": 5,
    "passed": 4,
    "failed": 1
  },
  "results": [
    {
      "test_name": "Verify homepage loads correctly",
      "status": "passed",
      "errors": [],
      "details": [...]
    }
  ]
}

Compatibility

• Requires n8n version 1.0.0 or later
• Tested with Playwright version 1.49.0
• Supports Windows, macOS, and Linux

System Requirements

• Node.js 18.10 or later
• Approximately 1GB disk space for browser binaries
• Additional system dependencies may be required for browser automation

Example AI Prompt for Test Generation

Use this prompt with your AI model to generate compatible test specifications:

Analyze the provided HTML and generate test cases in JSON format.

SUPPORTED ACTIONS:
Navigation: load_page, go_back, go_forward, reload_page
Interaction: click, fill_input, select_option, check_checkbox, uncheck_checkbox, hover, press_key, submit_form, scroll_to_element, double_click, right_click, drag_and_drop, focus, blur
Assertions: assert_element_present, assert_element_not_present, assert_element_visible, assert_text_present, assert_url_contains, assert_page_title, assert_element_count, assert_attribute, assert_element_enabled, assert_element_disabled, assert_checkbox_checked, assert_checkbox_unchecked
Wait: wait_for_element, wait
Utility: take_screenshot, upload_file, clear_input, execute_script, set_viewport
Dialog: accept_dialog, dismiss_dialog
Frame: switch_to_iframe, switch_to_main
Cookie: set_cookie, get_cookie, clear_cookies
Data Extraction: get_text, get_attribute

Format:
[
  {
    "test_name": "Test description",
    "steps": [
      {
        "action": "action_name",
        "selector": "CSS selector (if applicable)",
        "text": "Text to verify (if applicable)",
        "value": "Value to input (if applicable)",
        "key": "Key to press (if applicable)",
        "attribute": "Attribute name (if applicable)",
        "count": "Expected count (if applicable)",
        "timeout": "Timeout in ms (if applicable)",
        "description": "Step description"
      }
    ],
    "expected_result": "Expected outcome"
  }
]

Example:
[
  {
    "test_name": "Login test",
    "steps": [
      {"action": "load_page"},
      {"action": "fill_input", "selector": "#username", "value": "testuser"},
      {"action": "fill_input", "selector": "#password", "value": "password123"},
      {"action": "click", "selector": "#login-button"},
      {"action": "wait_for_element", "selector": ".dashboard", "timeout": 5000},
      {"action": "assert_url_contains", "text": "/dashboard"},
      {"action": "assert_element_visible", "selector": ".welcome-message"}
    ],
    "expected_result": "User successfully logs in and sees dashboard"
  }
]

Resources

• n8n community nodes documentation
• Playwright documentation
• npm package

Troubleshooting

Browser executable not found (local mode)

If browsers are not installed correctly:

1. Clean the installation:

rm -rf ~/.cache/ms-playwright
# or for Windows:
rmdir /s /q %USERPROFILE%\AppData\Local\ms-playwright

2. Reinstall the package through n8n's Community Nodes interface

Running in Docker (recommended: use Remote mode)

If you are running n8n in Docker and don't want to install browsers inside the container, switch to Remote Execution Mode and run a separate playwright-server container (see Remote Execution Setup above).

Remote connection refused

• Make sure the Playwright server container is running and reachable from the n8n container
• Check that the WebSocket URL matches the container name and port (e.g. ws://playwright-server:3000)
• Ensure both containers are on the same Docker network

License

MIT

Author

Modified from the original n8n-nodes-playwright by Mohamed Toema

Built for AI-powered test automation 🤖🎯