Intellyo

AI-powered test runner GUI for web applications. Generate and execute automated browser tests with a simple web interface.

Features

• Quick Test Generation: Select a test scenario and browser, click to generate and run
• Multiple Browsers: Chrome, Firefox, and Safari support via Playwright
• Real-time Output: WebSocket-powered live test output streaming
• Test Library: Browse and run existing YAML test files
• AI Integration: Use local Ollama models or cloud API for test generation
• Credential Management: Built-in support for test user credentials

Requirements

• Node.js 18+
• pnpm (or npm/yarn)
• IntelliTester CLI
• Playwright browsers installed
• (Optional) Ollama for local AI models

Installation

# Clone the repository
git clone https://github.com/server9-dev/intellyo.git
cd intellyo

# Install dependencies
pnpm install

# Build the project
pnpm build

# Install IntelliTester globally
npm install -g intellitester

# Install Playwright browsers
npx playwright install chromium firefox webkit
npx playwright install-deps

Quick Start

# Start the server
pnpm start

# Or for development with auto-rebuild
pnpm dev

Open http://localhost:4445 in your browser.

Configuration

Environment Variables

Create a .env file in the project root:

# Optional: API key for cloud AI providers
INTELLYO_API_KEY=your-api-key-here

# Test directory (default: ../intellitester-tests)
TEST_DIR=/path/to/your/tests

Test Credentials

Default test credentials can be configured in src/server.ts:

const CREDENTIALS = {
  user: { email: 'user@test.com', password: 'password123' },
  creator: { email: 'creator@test.com', password: 'password123' }
};

Usage

Quick Run (Generate & Run)

1. Select a test type (Login, Signup, Smoke Test, etc.)
2. Choose your browser (Chrome, Firefox, Safari)
3. Set the base URL for your application
4. Toggle "Show browser window" for visible testing
5. Click "Generate & Run"

Existing Tests

1. Browse the test list in the "Existing Tests" panel
2. Use the search box to filter tests
3. Select a test and click "Run Selected Test"

Test Scenarios

Scenario	Description
Login Flow	Test user authentication
Signup Flow	Test user registration
Send Message	Test messaging functionality
Send Media	Test file/media uploads
View Profile	Test user profile pages
Checkout	Test purchase flow
Settings	Test user settings pages
Admin Dashboard	Test admin functionality
Smoke Test	Quick test of all major pages

YAML Test Format

Tests are written in YAML format compatible with IntelliTester:

name: My Test
platform: web

config:
  web:
    baseUrl: http://localhost:3000
    headless: false

variables:
  testEmail: user@test.com
  testPassword: password123

steps:
  - type: navigate
    value: /login

  - type: input
    target:
      css: "input[type='email']"
    value: "{{testEmail}}"

  - type: tap
    target:
      css: "button[type='submit']"

  - type: screenshot
    name: result.png

API Endpoints

Endpoint	Method	Description
/api/scenarios	GET	List available test scenarios
/api/tests	GET	List existing test files
/api/browsers	GET	List available browsers
/api/models	GET	List available Ollama models
/api/generate	POST	Generate a new test
/api/run	POST	Run a test file

Development

# Install dependencies
pnpm install

# Run in development mode (watches for changes)
pnpm dev

# Build for production
pnpm build

# Start production server
pnpm start

# Type check
pnpm typecheck

Project Structure

intellyo/
├── src/
│   └── server.ts      # Express server with WebSocket support
├── public/
│   └── index.html     # Frontend UI
├── dist/              # Compiled JavaScript (generated)
├── package.json
├── tsconfig.json
└── README.md

Troubleshooting

Browser not found

# Install Playwright browsers
npx playwright install chromium firefox webkit
npx playwright install-deps

Port already in use

# Kill process on port 4445
fuser -k 4445/tcp

Ollama not available

The app works without Ollama - test generation uses predefined templates. For AI-powered generation, install Ollama and pull a model:

ollama pull qwen2.5-coder:7b

License

MIT

Contributing

Contributions are welcome! Please open an issue or submit a pull request.