playwright-node-sdk

Run Playwright Tests on TestMu AI (Formerly LambdaTest) with the Playwright Node SDK

The Playwright Node SDK runs an existing Playwright Test suite on the TestMu AI cloud without code changes: no connect() calls, no cloud-specific fixtures and no capabilities in project names. The specs, fixtures and playwright.config.ts in this folder are plain Playwright, and lambdatest.yml decides where they run.

What this sample covers

Spec	What it shows
tests/search.spec.ts	Data-driven tests: one test per row in test-data/searches.json
tests/checkout.spec.ts	Page objects; a worker fixture that registers a shop account once per worker over HTTP; logging in; reusing the signed-in session cookies to add to cart and place an order
tests/files.spec.ts	Uploading a file from your machine and downloading a file back to it
tests/browser.spec.ts	Alert, confirm and prompt dialogs; an editor inside an iframe; two tabs sharing one cart
tests/tunnel.spec.ts	Testing an app that runs only on your machine, through the TestMu AI tunnel

playwright-node-sdk/
├── lambdatest.yml          # credentials, platforms, capabilities, tunnel
├── playwright.config.ts    # plain Playwright config (timeouts, reporters, local web server)
├── fixtures/test.ts        # page-object fixtures and the per-worker account fixture
├── pages/                  # page objects
├── test-data/              # search terms, account and address data, upload file
├── local-app/server.js     # small local app for the tunnel test
└── tests/

Prerequisites

• Node.js 18 or newer, and npm
• A TestMu AI (Formerly LambdaTest) account with your username and access key

Setup

Clone the repo and install the dependencies of this sample:

git clone https://github.com/LambdaTest/playwright-sample && cd playwright-sample/playwright-node-sdk
npm install

Add your username and access key to lambdatest.yml. Don't commit your real access key.

user: "YOUR_LAMBDATEST_USERNAME"
accessKey: "YOUR_LAMBDATEST_ACCESS_KEY"

Run tests

npm test

This runs npx playwright-node-sdk playwright test, which takes the same arguments as npx playwright test. For example, npm test -- tests/search.spec.ts runs one spec.

Every test runs on each platform in lambdatest.yml: Chrome on Windows 11, WebKit on macOS Sonoma, and a real Android device. The SDK marks each session passed or failed and names it after the test. View the results on your TestMu AI dashboard under the build name from lambdatest.yml.

To run the same suite on your own machine instead, install the browser once and use test:local:

npx playwright install chromium
npm run test:local

Test an app on your machine (tunnel)

1. Set tunnel: true under tunnelSettings in lambdatest.yml.
2. Run:

npm run test:tunnel

Playwright starts the app in local-app/ on http://localhost:4173 (the webServer block in playwright.config.ts). The SDK starts the tunnel before the tests and stops it afterwards, so the cloud browsers can open localhost. The tests find the tunnel by tunnelName: give each run that happens at the same time its own name.

Configure platforms and capabilities

Add or change entries under platforms:

platforms:
  - platform: "Windows 11"
    browserName: "chrome"       # chrome, MicrosoftEdge, pw-chromium, pw-firefox, pw-webkit
    browserVersion: "latest"
  - platformName: "android"     # real Android device
    deviceName: "Galaxy.*"
    platformVersion: ".*"
    isRealMobile: true
    browserName: "chrome"
    playwrightConfigOptions:
      testMatch: ["**/search.spec.ts"]   # run only some specs on this platform

• Top-level keys such as video, console, network, geoLocation, timezone and resolution apply to every platform. ltOptions under a platform overrides them for that platform only.
• Set timeouts in playwright.config.ts, not in lambdatest.yml.
• parallelsPerPlatform × the number of platforms is the number of tests that run at once. Keep it within your plan's parallel limit.
• Real Android devices accept Playwright up to 1.61.1 today, so this sample pins 1.61.1 for every platform. If you test only desktop browsers, you can move to a newer version; keep playwright and @playwright/test on the same version.

Run npx playwright-node-sdk init to generate a full template, and see the SDK README for all tunnel settings and preserveProjects.

To keep several configurations, point the SDK at another file:

LAMBDATEST_YML_FILE=lambdatest.ci.yml npm test