mirror of
https://github.com/proffesor-for-testing/agentic-qe.git
synced 2026-09-19 08:45:47 +08:00
494cb11c1f
Added complete Playwright-based E2E testing infrastructure with 36 tests covering functional, responsive, accessibility, and performance testing. **Test Infrastructure:** - Playwright config with 7 browser/device configurations - Page Object Model for maintainable test code - npm scripts for various test execution modes **Test Suites (36 tests total):** - homepage.spec.ts: 14 functional tests (navigation, hero, stats, CTAs) - responsive.spec.ts: 5 responsive design tests (mobile, tablet, desktop) - accessibility.spec.ts: 10 WCAG 2.1 AA compliance tests - performance.spec.ts: 7 performance and Web Vitals tests **Documentation:** - E2E-TEST-PLAN.md: Comprehensive 15-section test strategy (700+ lines) - README.md: Complete usage guide (300+ lines) - QUICKSTART.md: 5-minute setup guide - DELIVERABLES.md: Detailed deliverables summary **Key Features:** - Based on real content crawled from live site - TypeScript for type safety - axe-core integration for accessibility testing - Screenshot and video capture on failure - CI/CD ready with GitHub Actions example - Cross-browser: Chrome, Firefox, Safari, Edge, Mobile - Responsive: Mobile (iPhone 12), Tablet (iPad), Desktop **Test Coverage:** ✅ All 7 navigation items ✅ Hero section with community stats (100K+ members) ✅ 4 explore cards (About, Impact, Community, Projects) ✅ 7 Paths of Impact ✅ "What is Agentic AI" section ✅ Community stats (130K+ Reddit, 52K+ LinkedIn, 3K+ Discord) ✅ SEO metadata and Open Graph tags ✅ HTTPS enforcement ✅ Footer and copyright 🤖 Generated with Claude Code (https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com>
3.5 KiB
3.5 KiB
Quick Start Guide - Agentics.org E2E Tests
Get up and running with the E2E test suite in 5 minutes!
Prerequisites
- Node.js 18+ installed
- npm or yarn
Installation (First Time Only)
# Navigate to the test directory
cd agentics
# Install dependencies
npm install
# Install Playwright browsers
npx playwright install
Run Your First Test
# Run all tests (headless)
npm test
# Run with visible browser (recommended for first time)
npm run test:headed
View Test Results
After running tests:
# Open HTML report
npm run test:report
Common Commands
# Run in interactive UI mode (best for development)
npm run test:ui
# Run only Chrome tests
npm run test:chromium
# Run only Firefox tests
npm run test:firefox
# Run mobile tests
npm run test:mobile
# Debug a specific test
npx playwright test tests/homepage.spec.ts --debug
Understanding the Output
Successful Test Run
✓ tests/homepage.spec.ts:26:3 › TC-001: Homepage should load
✓ tests/homepage.spec.ts:52:3 › TC-002: SEO and Metadata
...
14 passed (12.5s)
Failed Test
✗ tests/homepage.spec.ts:26:3 › TC-001: Homepage should load
Error: expect(received).toBe(expected)
Screenshot: test-results/.../screenshot.png
File Structure Quick Reference
agentics/
├── tests/ # Your test files
│ ├── homepage.spec.ts # Main functional tests
│ ├── responsive.spec.ts # Mobile/tablet tests
│ ├── accessibility.spec.ts # WCAG compliance
│ └── performance.spec.ts # Speed and Web Vitals
├── pages/ # Page Object Models
│ └── HomePage.ts # Homepage interactions
└── playwright.config.ts # Test configuration
What Gets Tested?
✅ Functional (14 tests)
- Navigation, hero section, community stats, CTAs
✅ Responsive (5 tests)
- Mobile, tablet, desktop layouts
✅ Accessibility (10 tests)
- WCAG 2.1 AA compliance, keyboard nav
✅ Performance (7 tests)
- Page speed, Core Web Vitals
Total: 36 automated tests
Writing Your First Test
import { test, expect } from '@playwright/test';
import { HomePage } from '../pages/HomePage';
test('my first test', async ({ page }) => {
const homePage = new HomePage(page);
await homePage.goto();
// Verify something
await expect(homePage.logo).toBeVisible();
});
Debugging Tips
- Use headed mode:
npm run test:headed - Use UI mode:
npm run test:ui(interactive) - Debug single test:
npx playwright test tests/homepage.spec.ts --debug - View trace: Look in
test-results/folder after failed test - Check screenshot: Failed tests save screenshots automatically
Next Steps
- ✅ Read
README.mdfor full documentation - ✅ Check
E2E-TEST-PLAN.mdfor test strategy - ✅ Review
DELIVERABLES.mdfor what's included - ✅ Explore test files in
tests/folder - ✅ Modify tests as website changes
Need Help?
- Playwright Docs: https://playwright.dev/
- Test Plan: See
E2E-TEST-PLAN.md - Full Guide: See
README.md
CI/CD Integration
Add to .github/workflows/e2e.yml:
name: E2E Tests
on: [push, pull_request]
jobs:
test:
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v3
- uses: actions/setup-node@v3
- run: cd agentics && npm ci
- run: cd agentics && npx playwright install --with-deps
- run: cd agentics && npm test
That's it! You're ready to run E2E tests for Agentics.org 🚀