Selenium With TestCafe

---

Run TestCafe tests on the TestMu AI cloud grid. This guide covers setup, running a sample test, configuring capabilities, and testing locally hosted pages.

npm package

Visit the TestMu AI npm package for TestCafe for full details.

Prerequisites

---

Complete these steps before integrating TestCafe with TestMu AI.

1. Create a TestMu AI account and get your username and access key from the dashboard.
2. Add the dependency "testcafe": "<testcafe_version>" in your package.json file.

Step 1: Clone the Sample Project

---

Install the TestMu AI npm plugin for TestCafe.

npm install testcafe-browser-provider-lambdatest

Step 2: Set Your Credentials

---

Set your TestMu AI username and access key as environment variables.

macOS / Linux

export LT_USERNAME="undefined"
export LT_ACCESS_KEY="undefined"

Windows

set LT_USERNAME="undefined"
set LT_ACCESS_KEY="undefined"

Step 3: Configure Your Test Capabilities

---

Configure your tests using environment variables for resolution, build name, and other settings.

VARIABLE	DESCRIPTION
LT_TEST_NAME	Specify a name for your test on TestMu AI Selenium Grid.
LT_BUILD	Specify a name for your build on TestMu AI Selenium Grid.
LT_CAPABILITY_PATH	Specify the path of a JSON file that declares additional Desired Capabilities. "[email protected]:Windows 8.1" : { "network" : true, "visual" : true, "timezone" : "UTC+11:00" } Here, [email protected]:Windows 8.1 represents browser alias.
LT_RESOLUTION	Perform cross browser testing on a specific screen resolution in the ${width}x${height} format.
LT_LOGFILE	View detailed logs of your automation script execution. Provide a specific path to this file. If you don't provide a path, the logs save in your present working directory by the filename: tunnel.log.
LT_VERBOSE	Set flag for verbose to either true or false.
LT_PROXY_HOST	A required variable if you perform cross browser testing through proxy. Declare the hostname/IP of proxy.
LT_PROXY_PORT	Port of the proxy. Default value for proxy port is 3128.
LT_PROXY_USER	Username for connecting to proxy. Required for using 'proxypass'.
LT_PROXY_PASS	Password for the USERNAME option.
LT_TUNNEL_NAME	Human readable tunnel identifier (Name of the tunnel).
LT_DIR	Path of the local folder you want to test.
LT_SELENIUM_VERSION	Browser specific capability.
LT_CONSOLE	Set flag as true or false.
LT_NETWORK	Set flag as true or false.
LT_VIDEO	Set flag as true or false.
LT_SCREENSHOT	Set flag as true or false.
LT_TIMEZONE	Configure tests to run on a custom time zone.
LT_TUNNEL_NUMBER	Number of tunnels to spawn at a time.
LOAD_BALANCED_MODE	Load Balanced Mode in TestCafe.

tip

Generate capabilities for your test requirements with the Capabilities Generator.

Step 4: Run the Test

---

Execute the test using the following command.

Single test:

$ testcafe "lambdatest:[email protected]:Windows 8" 'path/to/test/file.js'

Parallel tests:

$ testcafe "lambdatest:[email protected]:Windows 8","lambdatest:[email protected]:Windows 10" "path/to/test/file.js"

Using the API:

testCafe
    .createRunner()
    .src('path/to/test/file.js')
    .browsers('lambdatest:[email protected]:Windows 8')
    .run();

To list all available browser aliases:

testcafe -b lambdatest

Step 5: View Your Results

---

After running the test, view your results on the TestMu AI Automation Dashboard.

The dashboard provides:

• Video recordings of each test session
• Screenshots captured at each step
• Console logs from the browser
• Network logs for debugging
• Detailed command logs

Run TestCafe Tests Using Agent Skills

---

Use AI coding assistants to generate and run TestCafe tests with the TestMu AI Agent Skill.

The testcafe-skill is part of TestMu AI Agent Skills - structured packages that teach AI coding assistants how to write production-grade test automation.

Install the skill:

git clone https://github.com/LambdaTest/agent-skills.git
cp -r agent-skills/testcafe-skill .claude/skills/

# For Cursor / Copilot
cp -r agent-skills/testcafe-skill .cursor/skills/

tip

Install all available framework skills at once by cloning the repository directly into your tool's skills directory (e.g., .claude/skills/, .cursor/skills/).