playwright-pro（Playwright 专业版）

Playwright Pro

Tier: POWERFUL Category: Engineering / Testing Maintainer: Claude Skills Team

Overview

Production-grade end-to-end testing with Playwright. Generate tests from user stories, implement the Page Object pattern for maintainability, apply the correct locator strategy for resilient tests, diagnose and fix flaky tests, migrate from Cypress or Selenium, integrate with CI/CD, run visual regression tests, and perform accessibility audits. Enforces the 10 golden rules that eliminate 90% of E2E test failures.

Sub-Skills

This skill uses compound sub-skill architecture. Each sub-skill in skills/ handles a specific workflow:

Sub-Skill Flow

Init ──> Generate ──> Review ──> Fix (if needed)
                         │
                    Coverage ──> Generate (fill gaps)
                         │
                    Report ──> BrowserStack / TestRail

Typical lifecycle: Init sets up the project, Generate creates tests from stories, Review audits quality, Fix resolves failures, Coverage identifies gaps that feed back into Generate, and Report/BrowserStack/TestRail handle reporting and integration.

Scripts

Keywords

Playwright, E2E testing, end-to-end testing, page objects, flaky tests, test generation, Cypress migration, Selenium migration, visual regression, accessibility testing, CI integration

10 Golden Rules

These rules are non-negotiable. Following them eliminates 90% of E2E test failures.

1. getByRole() over CSS/XPath — resilient to markup changes
2. Never page.waitForTimeout() — use web-first assertions instead
3. expect(locator) auto-retries; expect(await locator.textContent()) does NOT
4. Isolate every test — no shared state between tests
5. baseURL in config — zero hardcoded URLs in tests
6. Retries: 2 in CI, 0 locally — retries mask flakiness in dev
7. Traces: 'on-first-retry' — rich debugging without slowdown
8. Fixtures over globals — test.extend() for shared setup
9. One behavior per test — multiple related assertions are fine
10. Mock external services only — never mock your own app

Locator Priority (Most to Least Preferred)

1. getByRole('button', { name: 'Submit' })     — semantic, accessible
2. getByLabel('Email address')                   — form fields with labels
3. getByText('Welcome back')                     — visible text content
4. getByPlaceholder('Enter your email')          — inputs with placeholder
5. getByTestId('submit-button')                  — when no semantic option exists
6. page.locator('.submit-btn')                   — CSS as last resort
7. page.locator('//button[@type="submit"]')      — XPath: avoid entirely

Why This Order Matters

// FRAGILE: breaks when CSS class changes
await page.locator('.btn-primary-lg').click();

// FRAGILE: breaks when DOM structure changes
await page.locator('div > form > button:nth-child(2)').click();

// RESILIENT: survives refactors, tests what users see
await page.getByRole('button', { name: 'Create account' }).click();

Configuration

playwright.config.ts

import { defineConfig, devices } from '@playwright/test';

export default defineConfig({
  testDir: './tests/e2e',
  fullyParallel: true,
  forbidOnly: !!process.env.CI,
  retries: process.env.CI ? 2 : 0,
  workers: process.env.CI ? 4 : undefined,
  reporter: process.env.CI
    ? [['html'], ['github'], ['json', { outputFile: 'test-results.json' }]]
    : [['html']],

  use: {
    baseURL: process.env.BASE_URL || 'http://localhost:3000',
    trace: 'on-first-retry',
    screenshot: 'only-on-failure',
    video: 'retain-on-failure',
  },

  projects: [
    // Auth setup: runs once, shares state with all tests
    { name: 'setup', testMatch: /.*\.setup\.ts/ },

    {
      name: 'chromium',
      use: { ...devices['Desktop Chrome'] },
      dependencies: ['setup'],
    },
    {
      name: 'firefox',
      use: { ...devices['Desktop Firefox'] },
      dependencies: ['setup'],
    },
    {
      name: 'mobile-chrome',
      use: { ...devices['Pixel 5'] },
      dependencies: ['setup'],
    },
  ],

  webServer: {
    command: 'pnpm dev',
    url: 'http://localhost:3000',
    reuseExistingServer: !process.env.CI,
    timeout: 30000,
  },
});

Page Object Pattern

// pages/login.page.ts
import { type Page, type Locator, expect } from '@playwright/test';

export class LoginPage {
  readonly page: Page;
  readonly emailInput: Locator;
  readonly passwordInput: Locator;
  readonly submitButton: Locator;
  readonly errorMessage: Locator;
  readonly forgotPasswordLink: Locator;

  constructor(page: Page) {
    this.page = page;
    this.emailInput = page.getByLabel('Email address');
    this.passwordInput = page.getByLabel('Password');
    this.submitButton = page.getByRole('button', { name: 'Sign in' });
    this.errorMessage = page.getByRole('alert');
    this.forgotPasswordLink = page.getByRole('link', { name: 'Forgot password?' });
  }

  async goto() {
    await this.page.goto('/login');
  }

  async login(email: string, password: string) {
    await this.emailInput.fill(email);
    await this.passwordInput.fill(password);
    await this.submitButton.click();
  }

  async expectError(message: string) {
    await expect(this.errorMessage).toContainText(message);
  }

  async expectRedirectToDashboard() {
    await expect(this.page).toHaveURL(/\/dashboard/);
  }
}

// pages/dashboard.page.ts
import { type Page, type Locator, expect } from '@playwright/test';

export class DashboardPage {
  readonly page: Page;
  readonly heading: Locator;
  readonly projectList: Locator;
  readonly createProjectButton: Locator;

  constructor(page: Page) {
    this.page = page;
    this.heading = page.getByRole('heading', { name: 'Dashboard' });
    this.projectList = page.getByRole('list', { name: 'Projects' });
    this.createProjectButton = page.getByRole('button', { name: 'New project' });
  }

  async expectLoaded() {
    await expect(this.heading).toBeVisible();
  }

  async getProjectCount() {
    return this.projectList.getByRole('listitem').count();
  }

  async createProject(name: string) {
    await this.createProjectButton.click();
    await this.page.getByLabel('Project name').fill(name);
    await this.page.getByRole('button', { name: 'Create' }).click();
  }
}

Test Generation from User Stories

Given a user story, generate tests following this pattern:

// tests/e2e/auth/login.spec.ts
import { test, expect } from '@playwright/test';
import { LoginPage } from '../../pages/login.page';
import { DashboardPage } from '../../pages/dashboard.page';

test.describe('User Login', () => {
  let loginPage: LoginPage;

  test.beforeEach(async ({ page }) => {
    loginPage = new LoginPage(page);
    await loginPage.goto();
  });

  test('successful login redirects to dashboard', async ({ page }) => {
    await loginPage.login('user@example.com', 'password123');
    const dashboard = new DashboardPage(page);
    await dashboard.expectLoaded();
  });

  test('shows error for invalid credentials', async () => {
    await loginPage.login('user@example.com', 'wrongpassword');
    await loginPage.expectError('Invalid email or password');
  });

  test('shows validation error for empty email', async () => {
    await loginPage.login('', 'password123');
    await loginPage.expectError('Email is required');
  });

  test('shows validation error for invalid email format', async () => {
    await loginPage.login('not-an-email', 'password123');
    await loginPage.expectError('Enter a valid email');
  });

  test('forgot password link navigates to reset page', async ({ page }) => {
    await loginPage.forgotPasswordLink.click();
    await expect(page).toHaveURL(/\/forgot-password/);
  });
});

Authentication Setup (Shared State)

// tests/e2e/auth.setup.ts
import { test as setup, expect } from '@playwright/test';

const authFile = 'playwright/.auth/user.json';

setup('authenticate as test user', async ({ page }) => {
  await page.goto('/login');
  await page.getByLabel('Email address').fill('test@example.com');
  await page.getByLabel('Password').fill('password123');
  await page.getByRole('button', { name: 'Sign in' }).click();

  // Wait for redirect to confirm login succeeded
  await expect(page).toHaveURL(/\/dashboard/);

  // Save authentication state for reuse
  await page.context().storageState({ path: authFile });
});

Flaky Test Diagnosis

Common Causes and Fixes

Diagnosis Commands

# Run with trace on every test (not just retries)
npx playwright test --trace on

# Run a specific flaky test 10 times
for i in $(seq 1 10); do npx playwright test tests/e2e/checkout.spec.ts; done

# Show test timeline
npx playwright test --trace on
npx playwright show-trace test-results/*/trace.zip

# Run in headed mode for visual debugging
npx playwright test --headed --retries 0

# Debug a specific test interactively
npx playwright test --debug tests/e2e/checkout.spec.ts

Cypress to Playwright Migration

CI Integration

GitHub Actions

name: E2E Tests
on: [push, pull_request]

jobs:
  e2e:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v4
      - uses: pnpm/action-setup@v4
      - uses: actions/setup-node@v4
        with:
          node-version: 20
          cache: 'pnpm'
      - run: pnpm install --frozen-lockfile
      - run: pnpm exec playwright install --with-deps chromium

      - run: pnpm build  # build the app first
      - run: pnpm exec playwright test
        env:
          BASE_URL: http://localhost:3000

      - uses: actions/upload-artifact@v4
        if: failure()
        with:
          name: playwright-report
          path: playwright-report/
          retention-days: 7

Visual Regression Testing

// tests/e2e/visual/dashboard.spec.ts
import { test, expect } from '@playwright/test';

test('dashboard matches visual snapshot', async ({ page }) => {
  await page.goto('/dashboard');
  await page.waitForLoadState('networkidle');

  // Full page screenshot comparison
  await expect(page).toHaveScreenshot('dashboard.png', {
    maxDiffPixels: 50,  // allow small rendering differences
  });
});

test('project card component snapshot', async ({ page }) => {
  await page.goto('/dashboard');
  const card = page.getByTestId('project-card').first();

  await expect(card).toHaveScreenshot('project-card.png', {
    maxDiffPixelRatio: 0.01,
  });
});

# Generate/update baseline screenshots
npx playwright test --update-snapshots

# Run visual comparison
npx playwright test tests/e2e/visual/

Accessibility Testing

import { test, expect } from '@playwright/test';
import AxeBuilder from '@axe-core/playwright';

test('login page has no accessibility violations', async ({ page }) => {
  await page.goto('/login');

  const results = await new AxeBuilder({ page })
    .withTags(['wcag2a', 'wcag2aa', 'wcag21a', 'wcag21aa'])
    .analyze();

  expect(results.violations).toEqual([]);
});

test('dashboard meets WCAG AA standards', async ({ page }) => {
  await page.goto('/dashboard');

  const results = await new AxeBuilder({ page })
    .exclude('.third-party-widget')  // exclude elements you don't control
    .analyze();

  // Log violations for debugging
  for (const violation of results.violations) {
    console.log(`${violation.impact}: ${violation.description}`);
    for (const node of violation.nodes) {
      console.log(`  - ${node.html}`);
    }
  }

  expect(results.violations.filter(v => v.impact === 'critical')).toEqual([]);
});

Common Pitfalls

• page.waitForTimeout(N) — the single most common cause of flaky tests; use web-first assertions
• CSS selectors as primary strategy — breaks on every refactor; use role/label/text locators
• Shared state between tests — one test's data pollutes another; isolate with proper setup/teardown
• No trace configuration — debugging CI failures without traces wastes hours; enable on-first-retry
• Testing third-party services — mock external APIs; only test your own application
• Running all browsers in development — test Chromium locally, full matrix in CI only
• No page objects — duplicate locators across tests create maintenance nightmares

Best Practices

1. Page Object per page/component — centralize locators, expose user-intent methods
2. Web-first assertions everywhere — expect(locator).toBeVisible() auto-retries, waitForTimeout does not
3. Auth as a setup project — authenticate once, reuse storageState across all tests
4. One behavior per test — keeps failures isolated and test names meaningful
5. Run in CI with --retries 2 — but investigate any test that needs retries locally
6. Trace + screenshot on failure — upload as CI artifacts for post-mortem debugging
7. Visual regression for critical UI — catch unintended visual changes automatically
8. Accessibility tests in the suite — WCAG compliance as a regression gate, not an afterthought

Troubleshooting

Success Criteria

• Flaky test rate below 2% — measured over a rolling 7-day window across all CI runs; any test exceeding 5% flake rate is quarantined and fixed within 48 hours
• E2E suite completes within 10 minutes — full cross-browser matrix (Chromium, Firefox, mobile) on CI with 4 parallel workers; alert if wall-clock time exceeds threshold
• 100% of critical user journeys covered — login, checkout, onboarding, and core CRUD workflows each have dedicated spec files with happy-path and primary error-path tests
• Zero waitForTimeout() calls in the codebase — enforced via ESLint rule or grep-based CI check; all waits use web-first assertions
• Page Object coverage for every tested page — no raw locators in spec files; all element access goes through page object classes with user-intent method names
• WCAG AA compliance gate passing — accessibility tests run on every PR; zero critical or serious violations allowed to merge
• Visual regression baselines reviewed on every UI PR — screenshot diffs attached to PR as artifacts; baseline updates require explicit reviewer approval

Scope & Limitations

This skill covers:

• End-to-end test authoring, organization, and maintenance with Playwright
• Page Object Model architecture and locator strategy best practices
• Flaky test diagnosis, CI integration, and trace-based debugging
• Visual regression testing and WCAG accessibility auditing

This skill does NOT cover:

• Unit testing or component testing in isolation (see engineering/testing-strategy for test pyramid guidance)
• API contract testing or load/performance testing (see api-test-suite-builder for API-focused testing)
• Test data management, database seeding, or factory patterns for test fixtures
• Mobile native app testing (Appium, Detox); this skill targets web browsers only

Integration Points