Run Selenium Tests Using Agent Skills

---

Instead of manually setting up your Selenium project, use Agent Skills to let your AI coding assistant generate production-ready test automation for you. The selenium-skill is part of TestMu AI Agent Skills - structured packages that teach AI assistants how to write production-grade test code.

What Are Agent Skills?

---

Self-contained packages of instructions, code patterns, and configurations for specific testing frameworks.

Once installed, your AI assistant works like a senior QA automation architect with knowledge of:

• Proper project structure for each framework
• Correct dependency versions and configurations
• Both local and TestMu AI cloud execution patterns
• Common pitfalls and debugging approaches
• CI/CD integration with GitHub Actions

Prerequisites

---

Make sure you have the following set up before you start.

1. Create a TestMu AI account if you don't have one.
2. Get your Username and Access Key from the TestMu AI Dashboard.
3. Install an AI coding assistant: Claude Code, GitHub Copilot, Cursor, Gemini CLI, or any compatible tool.

Step 1: Install the Selenium Agent Skill

---

Clone the repository and copy the Selenium skill into your AI tool's skills directory.

git clone https://github.com/LambdaTest/agent-skills.git

# For Claude Code
cp -r agent-skills/selenium-skill .claude/skills/

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

# For GitHub Copilot
cp -r agent-skills/selenium-skill .github/skills/

# For Gemini CLI
cp -r agent-skills/selenium-skill .gemini/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/).

Step 2: Set Your Credentials

---

Add your TestMu AI credentials as environment variables so the generated tests can authenticate with the grid.

Visit the TestMu AI Dashboard, navigate to the left sidebar, and select Credentials. Copy your Username and Access Key, then run:

export LT_USERNAME="YOUR_LAMBDATEST_USERNAME"
export LT_ACCESS_KEY="YOUR_LAMBDATEST_ACCESS_KEY"

Step 3: Run Your First Test

---

Ask your AI assistant to write tests in natural language.

"Write Selenium tests for the login page and run them on TestMu AI cloud using Chrome and Firefox"

"Set up a Selenium Java project with TestNG, Page Object Model, and run tests on TestMu AI"

"Execute Selenium tests across 5 browsers in parallel on TestMu AI, tag the build as 'release-v1.0'"

The Agent Skill automatically handles:

1. Project setup - Creates the right folder structure, dependencies, and config files
2. Language detection - Maps your request to the correct language (Java, Python, JavaScript, C#, Ruby, or PHP)
3. Execution target - Configures local or TestMu AI cloud execution based on your request
4. Best practices - Uses explicit waits, proper locator strategies, and Page Object Model patterns

What's Included in the Selenium Skill

---

A breakdown of everything the skill package contains.

selenium-skill/
├── SKILL.md
└── reference/
    ├── playbook.md
    └── advanced-patterns.md

Section	Details
Project Setup	Dependencies, versions, configuration files, project structure
Core Patterns	Complete, runnable code examples with best practices
Cloud Integration	TestMu AI RemoteWebDriver and capabilities configuration
CI/CD Integration	GitHub Actions workflows with reporting and parallel execution
Debugging Guide	11+ common exceptions mapped to solutions
Best Practices	14+ actionable recommendations for production code

Supported Languages and Frameworks

---

The Selenium Agent Skill supports 6 languages. Each framework below also has its own dedicated Agent Skill with framework-specific patterns.

Java

Framework	Agent Skill	Documentation
TestNG	testng-skill	TestNG with Selenium
JUnit 5	junit-5-skill	JUnit with Selenium
Cucumber	cucumber-skill	Cucumber with Selenium
Selenide	selenide-skill	Selenide Framework
Gauge	gauge-skill	Gauge Framework
Geb	geb-skill	Geb Framework
Serenity BDD	serenity-bdd-skill	Serenity BDD

JavaScript

Framework	Agent Skill	Documentation
Mocha	mocha-skill	Mocha with Selenium
Jest	jest-skill	Jest with Selenium
WebdriverIO	webdriverio-skill	WebdriverIO
Nightwatch.js	nightwatchjs-skill	Nightwatch.js
Protractor	protractor-skill	Protractor
TestCafe	testcafe-skill	TestCafe
Jasmine	jasmine-skill	Jasmine with Karma
Karma	karma-skill	Karma Integration
CucumberJS	cucumber-skill	CucumberJS
NemoJS	nemojs-skill	NemoJS

C#

Framework	Agent Skill	Documentation
NUnit	nunit-skill	NUnit with Selenium
MSTest	mstest-skill	MSTest with Selenium
xUnit	xunit-skill	xUnit with Selenium
SpecFlow	specflow-skill	SpecFlow with Selenium

Python

Framework	Agent Skill	Documentation
pytest	pytest-skill	pytest with Selenium
unittest	unittest-skill	unittest
Robot Framework	robot-framework-skill	Robot Framework
Behave	behave-skill	Behave with Selenium
Lettuce	lettuce-skill	Lettuce with Selenium

PHP

Framework	Agent Skill	Documentation
PHPUnit	phpunit-skill	PHPUnit
Behat	behat-skill	Behat
Laravel Dusk	laravel-dusk-skill	Laravel Dusk
Codeception	codeception-skill	Codeception

Ruby

Framework	Agent Skill	Documentation
RSpec	rspec-skill	RSpec
Capybara	capybara-skill	Capybara
Test::Unit	testunit-skill	Test::Unit

Supported AI Tools

---

Pick the tool you use and install the skill to the corresponding path.

Tool	Type	Installation Path
Claude Code	CLI	.claude/skills/
GitHub Copilot	Extension	.github/skills/
Cursor	IDE	.cursor/skills/
Gemini CLI	CLI	.gemini/skills/
Codex CLI	CLI	.codex/skills/
OpenCode	CLI	.opencode/skills/
Claude.ai	Web	Settings > Features > Skills