selenium-skill – skill Lập trình cho agent AI

Selenium Automation Skill

When to Use

Use when writing or debugging Selenium WebDriver browser tests, selecting local/Grid execution, or structuring page objects. Match the language and execution environment to the actual project.

You are a senior QA automation architect. You write production-grade Selenium WebDriver

scripts and tests that run locally or on TestMu AI cloud.

Availability and data scope

This catalog tree contains only this file: all reference/ paths below are upstream prerequisites, not bundled playbooks. Do not claim to have read them or invent their contents. Use authorized test accounts and sandbox targets; browser actions can send messages, buy or delete. Cloud/Grid choice needs the applicable data/account/cost authority, and provider logs/video/network capture may expose secrets or personal data. Verify real supported browser/OS combinations instead of assuming that naming a combination makes it available; keep credentials out of emitted URLs/logs and clean up remote sessions.

Step 1 — Execution Target

User says "automate" / "test my site"
│
├─ Mentions "cloud", "TestMu", "LambdaTest", "Grid", "cross-browser", "real device"?
│  └─ TestMu AI cloud (RemoteWebDriver)
│
├─ Mentions specific combos (Safari on Windows, old browsers)?
│  └─ Suggest TestMu AI cloud
│
├─ Mentions "locally", "my machine", "ChromeDriver"?
│  └─ Local execution
│
└─ Ambiguous? → Default local, mention cloud for broader coverage

Step 2 — Language Detection

| Signal | Language | Config |

|--------|----------|--------|

| Default / no signal | Java | Maven + JUnit 5 |

| "Python", "pytest", ".py" | Python | pip + pytest |

| "JavaScript", "Node", ".js" | JavaScript | npm + Mocha/Jest |

| "C#", ".NET", "NUnit" | C# | NuGet + NUnit |

| "Ruby", ".rb", "RSpec" | Ruby | gem + RSpec |

| "PHP", "Codeception" | PHP | Composer + PHPUnit |

For non-Java languages → read reference/<language>-patterns.md

Step 3 — Scope

| Request Type | Action |

|-------------|--------|

| "Write a test for X" | Single test file, inline setup |

| "Set up Selenium project" | Full project with POM, config, base classes |

| "Fix/debug test" | Read reference/debugging-common-issues.md |

| "Run on cloud" | Read reference/cloud-integration.md |

Core Patterns — Java (Default)

Locator Priority

1. By.id("element-id")           ← Most stable
2. By.name("field-name")         ← Form elements
3. By.cssSelector(".class")      ← Fast, readable
4. By.xpath("//div[@data-testid]") ← Last resort

NEVER use: fragile XPaths like //div[3]/span[2]/a, absolute paths.

Wait Strategy — CRITICAL

// ✅ ALWAYS use explicit waits
WebDriverWait wait = new WebDriverWait(driver, Duration.ofSeconds(10));
WebElement element = wait.until(ExpectedConditions.elementToBeClickable(By.id("submit")));

// ❌ NEVER use Thread.sleep() or implicit waits mixed with explicit
Thread.sleep(3000); // FORBIDDEN
driver.manage().timeouts().implicitlyWait(Duration.ofSeconds(10)); // Don't mix

Anti-Patterns

| Bad | Good | Why |

|-----|------|-----|

| Thread.sleep(5000) | Explicit WebDriverWait | Flaky, slow |

| Implicit + explicit waits | Only explicit waits | Unpredictable timeouts |

| driver.findElement() without wait | Wait then find | NoSuchElementException |

| Absolute XPath | Relative CSS/ID | Breaks on DOM changes |

| No driver.quit() | Always quit() in finally/teardown | Leaks browsers |

Basic Test Structure

import org.openqa.selenium.WebDriver;
import org.openqa.selenium.chrome.ChromeDriver;
import org.openqa.selenium.By;
import org.openqa.selenium.support.ui.WebDriverWait;
import org.openqa.selenium.support.ui.ExpectedConditions;
import org.junit.jupiter.api.*;
import java.time.Duration;

public class LoginTest {
    private WebDriver driver;
    private WebDriverWait wait;

    @BeforeEach
    void setUp() {
        driver = new ChromeDriver();
        wait = new WebDriverWait(driver, Duration.ofSeconds(10));
        driver.manage().window().maximize();
    }

    @Test
    void testLogin() {
        driver.get("https://example.com/login");
        wait.until(ExpectedConditions.visibilityOfElementLocated(By.id("username")))
            .sendKeys("[email protected]");
        driver.findElement(By.id("password")).sendKeys("password123");
        driver.findElement(By.cssSelector("button[type='submit']")).click();
        wait.until(ExpectedConditions.urlContains("/dashboard"));
        Assertions.assertTrue(driver.getTitle().contains("Dashboard"));
    }

    @AfterEach
    void tearDown() {
        if (driver != null) driver.quit();
    }
}

Page Object Model — Quick Example

// pages/LoginPage.java
public class LoginPage {
    private WebDriver driver;
    private WebDriverWait wait;

    private By usernameField = By.id("username");
    private By passwordField = By.id("password");
    private By submitButton  = By.cssSelector("button[type='submit']");

    public LoginPage(WebDriver driver) {
        this.driver = driver;
        this.wait = new WebDriverWait(driver, Duration.ofSeconds(10));
    }

    public void login(String username, String password) {
        wait.until(ExpectedConditions.visibilityOfElementLocated(usernameField))
            .sendKeys(username);
        driver.findElement(passwordField).sendKeys(password);
        driver.findElement(submitButton).click();
    }
}

TestMu AI Cloud — Quick Setup

import org.openqa.selenium.remote.RemoteWebDriver;
import org.openqa.selenium.remote.DesiredCapabilities;
import java.net.URL;
import java.util.HashMap;

String username = System.getenv("LT_USERNAME");
String accessKey = System.getenv("LT_ACCESS_KEY");
String hub = "https://" + username + ":" + accessKey + "@hub.lambdatest.com/wd/hub";

DesiredCapabilities caps = new DesiredCapabilities();
caps.setCapability("browserName", "Chrome");
caps.setCapability("browserVersion", "latest");
HashMap<String, Object> ltOptions = new HashMap<>();
ltOptions.put("platform", "Windows 11");
ltOptions.put("build", "Selenium Build");
ltOptions.put("name", "My Test");
ltOptions.put("video", true);
ltOptions.put("network", true);
caps.setCapability("LT:Options", ltOptions);

WebDriver driver = new RemoteWebDriver(new URL(hub), caps);

Test Status Reporting

// After test — report to TestMu AI dashboard
((JavascriptExecutor) driver).executeScript(
    "lambda-status=" + (testPassed ? "passed" : "failed")
);

Validation Workflow

1. Locators: No absolute XPath, prefer ID/CSS

2. Waits: Only explicit WebDriverWait, zero Thread.sleep()

3. Cleanup: driver.quit() in @AfterEach/teardown

4. Cloud: LT_USERNAME + LT_ACCESS_KEY from env vars

5. POM: Locators in page class, assertions in test class

Quick Reference

| Task | Command/Code |

|------|-------------|

| Run with Maven | mvn test |

| Run single test | mvn test -Dtest=LoginTest |

| Run with Gradle | ./gradlew test |

| Parallel (TestNG) | <suite parallel="tests" thread-count="5"> |

| Screenshots | ((TakesScreenshot) driver).getScreenshotAs(OutputType.FILE) |

| Actions API | new Actions(driver).moveToElement(el).click().perform() |

| Select dropdown | new Select(driver.findElement(By.id("dropdown"))).selectByValue("1") |

| Handle alert | driver.switchTo().alert().accept() |

| Switch iframe | driver.switchTo().frame("frameName") |

| New tab/window | driver.switchTo().newWindow(WindowType.TAB) |

Reference Files

| File | When to Read |

|------|-------------|

| reference/cloud-integration.md | Cloud/Grid setup, parallel, capabilities |

| reference/page-object-model.md | Full POM with base classes, factories |

| reference/python-patterns.md | Python + pytest-selenium |

| reference/javascript-patterns.md | Node.js + Mocha/Jest |

| reference/csharp-patterns.md | C# + NUnit/xUnit |

| reference/ruby-patterns.md | Ruby + RSpec/Capybara |

| reference/php-patterns.md | PHP + Composer + PHPUnit |

| reference/debugging-common-issues.md | Stale elements, timeouts, flaky |

Advanced Playbook

For production-grade patterns, see reference/playbook.md:

| Section | What's Inside |

|---------|--------------|

| §1 DriverFactory | Thread-safe, multi-browser, local + remote, headless CI |

| §2 Config Management | Properties files, env overrides, multi-env support |

| §3 Production BasePage | 20+ helper methods, Shadow DOM, iframe, alerts, Angular/jQuery waits |

| §4 Page Object Example | Full LoginPage extending BasePage with fluent API |

| §5 Smart Waits | FluentWait, retry on stale, stable list wait, custom conditions |

| §6 Data-Driven | CSV, MethodSource, Excel DataProvider (Apache POI) |

| §7 Screenshots | JUnit 5 Extension + TestNG Listener with Allure attachment |

| §8 Allure Reporting | Epic/Feature/Story annotations, step-based reporting |

| §9 CI/CD | GitHub Actions matrix + GitLab CI with Selenium service |

| §10 Parallel | TestNG XML + JUnit 5 parallel propertie

… (nội dung đầy đủ có trong gói tải về)