Full Web App Testing in 10 Minutes: Zero to E2E Suite
Suneet Malhotra
Jan 1, 2026
Full Web App Testing in 10 Minutes: Zero to E2E Suite
As a Quality Engineering leader, I've seen teams spend weeks—sometimes months—building comprehensive E2E test suites. But what if you could go from zero coverage to a full test suite in just 10 minutes?
This is the story of how I used native Playwright AI agents (Planner, Generator, and Healer) to audit my entire portfolio website, generate 23 comprehensive tests covering critical user journeys, and achieve a 100% pass rate—all triggered by a single prompt.
💡 Key Insight: Modern AI agents can understand your application structure, identify critical user flows, write production-ready test code, and automatically fix failures—all without manual intervention.
TL;DR: Key Takeaways
- Time Savings: 2-3 weeks of manual work → 10 minutes with AI agents
- Coverage: Generated 23 comprehensive E2E tests covering all critical user journeys
- Success Rate: 100% pass rate on first execution
- Agents Used: Planner (audit), Generator (code), Healer (self-repair)
- Technology: Native Playwright AI agents with zero configuration
Why Is Traditional Test Creation So Slow?
Traditional Approach:
- 📋 Planning: 2-4 hours analyzing requirements and user journeys
- ✍️ Writing Tests: 8-16 hours coding test scenarios
- 🐛 Debugging: 4-8 hours fixing selectors and timing issues
- 🔄 Maintenance: Ongoing hours per week fixing broken tests
Total Time: 2-3 weeks for a comprehensive suite
The Problem: By the time your tests are ready, the application has already changed.
How Does the AI-Powered 10-Minute Workflow Work?
Using Playwright's native AI agents, I transformed this multi-week process into a 10-minute automated workflow:
What Happens in the 10-Minute Breakdown?
0:00 - 2:00: Planner Agent
The Planner agent crawls your entire web application, analyzing:
- All routes and navigation paths
- User interaction points
- Form submissions
- Critical business flows
- Edge cases and error states
Output: A comprehensive test plan identifying 20+ critical user journeys
# E2E Test Plan for suneetmalhotra.com
## Journey 1: First-Time Visitor Exploration
- Navigate to homepage
- Verify hero section loads
- Check navigation links
- Verify responsive design
## Journey 2: AI Tools Discovery
- Navigate to /tools
- Verify tool cards display
- Access demo pages
- Verify external links
## Journey 6: Navigation & Smart Navbar
- Test all navigation links
- Verify active states
- Test mobile menu
- Verify smooth scrolling
2:00 - 6:00: Generator Agent
The Generator agent converts the test plan into executable TypeScript code:
- Creates production-ready Playwright test files
- Uses best-practice selectors (
getByRole,getByLabel) - Includes proper waits and assertions
- Adds detailed logging for visibility
Output: 23 complete test files with 100% of the code written automatically
// Example: Generated test for First-Time Visitor Journey
import { test, expect } from '@playwright/test';
test.describe('Journey 1: First-Time Visitor Exploration', () => {
test('should load homepage and display hero section', async ({ page }) => {
console.log('🚀 Starting homepage verification...');
await page.goto('https://suneetmalhotra.com');
await page.waitForLoadState('networkidle');
console.log('📍 Verifying hero section...');
const hero = page.locator('h1').first();
await expect(hero).toBeVisible();
console.log('✅ Homepage loaded successfully');
});
});
6:00 - 9:00: Healer Agent
The Healer agent runs the test suite and automatically fixes any failures:
- Detects selector mismatches
- Analyzes screenshots and error traces
- Generates new locators
- Retries with corrected selectors
- Achieves 100% pass rate
Output: All 23 tests passing with zero manual intervention
9:00 - 10:00: Final Execution & Reporting
- Runs all tests in browser (headed mode for visibility)
- Generates comprehensive Allure reports
- Provides test summaries, statistics, and execution timeline
Output: Complete E2E coverage with detailed reporting
The Results: Verified Metrics
After executing the complete workflow, here's what was achieved:
| Metric | Result |
|---|---|
| Tests Generated | 23 comprehensive E2E tests |
| Pass Rate | 100% (all tests passing) |
| Total Time | ~10 minutes |
| User Journeys Covered | 3 P0 (Critical) journeys |
| Code Written | 100% automated |
| Manual Fixes Required | 0 |
Test Coverage Breakdown
Journey 1: First-Time Visitor Exploration (7 tests)
- Homepage loading and hero section
- Navigation links functionality
- Responsive design verification
- Profile image display
- Social links accessibility
Journey 2: AI Tools Discovery and Demo Access (7 tests)
- Tools page navigation
- Tool card display and interaction
- Demo page access
- External link verification
- GitHub repository links
Journey 6: Navigation and Smart Navbar (9 tests)
- All navigation links
- Active state indicators
- Mobile menu functionality
- Smooth scrolling behavior
- Route transitions
How It Works: The Three-Agent System
🤖 Planner Agent
Role: Strategic test planning
- Analyzes application structure
- Identifies critical user journeys
- Creates comprehensive test scenarios
- Documents expected outcomes
Tools: Browser exploration, route analysis, user flow mapping
✍️ Generator Agent
Role: Code generation
- Converts test plans to executable code
- Uses Playwright best practices
- Implements robust selectors
- Adds proper error handling
Tools: TypeScript generation, selector optimization, code formatting
🔧 Healer Agent
Role: Automatic test repair
- Detects and fixes selector failures
- Handles timing issues
- Manages dynamic content
- Ensures test stability
Tools: Failure analysis, locator regeneration, retry logic
The Single-Prompt Workflow
The entire process is triggered by a single, comprehensive prompt:
# Role: Senior QA Automation Engineer & AI Test Agent Orchestrator
# Goal: Execute complete end-to-end test automation workflow
I want to execute the complete Playwright Test Agents workflow
(Planner → Generator → Healer) to create, automate, and execute
P0 (Critical) E2E tests, then generate comprehensive Allure reports.
## Workflow Steps:
1. Planner: Generate comprehensive test plan
2. Generator: Create executable test files
3. Healer: Fix any failures automatically
4. Execute: Run tests in browser with visibility
5. Report: Generate Allure reports
Result: One prompt → Complete test suite → 100% pass rate
Why This Matters for QE Leaders
Speed to Coverage
- Traditional: 2-3 weeks to comprehensive coverage
- AI Agents: 10 minutes to full E2E suite
- Time Saved: 99% reduction in test creation time
Quality Assurance
- Automated Best Practices: Agents use industry-standard patterns
- Comprehensive Coverage: No critical flows missed
- Self-Healing: Tests adapt to UI changes automatically
Cost Efficiency
- Zero Manual Coding: 100% automated test generation
- Reduced Maintenance: Self-healing reduces ongoing costs
- Faster Feedback: Issues detected in minutes, not weeks
Try It Yourself
Experience the complete workflow on the demo page:
- Watch the 10-Minute Workflow Video: See the complete process from start to finish
- Review the Test Plan: Explore the comprehensive test scenarios
- Examine Generated Tests: Review the production-ready code
- View Allure Reports: See detailed execution results
Quick Start
# 1. Clone the repository
git clone https://github.com/SuneetMalhotra/professional-portfolio.git
cd professional-portfolio/qa-automation
# 2. Install dependencies
npm install
# 3. Run the complete workflow
npm run test:e2e
# 4. Generate Allure reports
npm run allure:auto
The Future of Test Automation
This 10-minute workflow represents a fundamental shift in how we approach test automation:
- From Manual to Automated: No more weeks of test writing
- From Reactive to Proactive: Tests generated before deployment
- From Brittle to Resilient: Self-healing tests adapt automatically
- From Expensive to Efficient: 99% reduction in time and cost
The question isn't whether AI will transform test automation—it's whether you're ready to embrace it.
Related Resources
- Live Demo: Interactive demonstration of the 10-minute workflow
- Workflow Execution Summary: Detailed metrics and results
- Demo Prompt: Copy-paste ready prompt for your own workflow
- GitHub Repository: Complete source code and test files
About the Author: Suneet Malhotra is an AI-Driven Quality Engineering Leader with 20+ years of experience in scaling complex platforms. This workflow was executed end-to-end and verified with real metrics.
Share this post
You Might Also Like
Building Shape Popper: A Kid-Friendly iOS Game with SwiftUI & Claude Code
Step-by-step guide on building Shape Popper, a high-performance SwiftUI game for kids using Claude Code. Learn about Canvas rendering, MVVM architecture, and iOS accessibility for ages 4-6.
Engineering LeadershipBuild a Project Management App with Claude Code in 15 Minutes
A step-by-step guide to building Flowstate with Claude Code, Next.js, and Supabase. Learn how to use /frontend-design for a custom Editorial Brutalism aesthetic and deploy a fully functional project management app in 15 minutes.
Agentic AIEverything in My Context Window Is an Instruction
This routine reads the open web, then commits to a live site with no human in the loop. Those two facts sit in the same context window, and the model has no way to tell them apart.
Quantitative TradingThe Edge I Assume Is Already Decaying
Wall Street spent this week arguing that AI is cutting the useful life of a trading edge from seven years to eighteen months. If that is even half right, it changes what a signal is worth.
Latest Blog Posts
Everything in My Context Window Is an Instruction
This routine reads the open web, then commits to a live site with no human in the loop. Those two facts sit in the same context window, and the model has no way to tell them apart.
The Edge I Assume Is Already Decaying
Wall Street spent this week arguing that AI is cutting the useful life of a trading edge from seven years to eighteen months. If that is even half right, it changes what a signal is worth.
The If Statement My Audit Never Read
On May 20 I published a rule for which steps of a routine are safe to run twice, and put my repo pull in the safest bucket. On July 9 that step failed. It never ran.
Related Tools & Demos
Multi-Model LLM Harness
One interface to call any AI model — capability routing, fallback chains, budgets, circuit breakers, and a quality feedback loop. A practical architecture pattern write-up.
Automated Trading System
Multi-engine trading platform with real-time risk management, regime-based strategy selection, and automated order execution.
View Source Code →Personal Health Analytics
Multi-modal health data platform integrating wearables, lab results, and lifestyle tracking with predictive habit modeling.
View Source Code →
Stay in the Loop
Get weekly insights on AI-driven QA, engineering leadership, and automation strategies.
No spam, ever. Unsubscribe anytime.