Package Exports

playwright-genie-ai
playwright-genie-ai/dist/index.js

This package does not declare an exports field, so the exports above have been automatically detected and optimized by JSPM instead. If any package subpath is missing, it is recommended to post an issue to the original package (playwright-genie-ai) to support the "exports" field. If that is not possible, create a JSPM override to customize the exports field for this package.

Readme

🧞 Playwright Genie AI

AI-powered Playwright test generator that creates test scripts from natural language scenarios.

🚀 Features

Natural Language Testing: Describe your test scenario in plain English
Multiple AI Models: Support for GPT-4, GPT-3.5, and local Ollama models
Automatic Test Generation: Creates complete Playwright .spec.ts files
Instant Execution: Option to run tests immediately after generation
Smart Selectors: Uses best practices for element selection
NPM CLI Tool: Easy to install and use anywhere

📦 Installation

Global Installation (Recommended)

npm install -g playwright-genie-ai

One-time Usage

npx playwright-genie-ai --url <url> --scenario "<scenario>"

Prerequisites

Node.js 16.0.0 or higher
npm 7.0.0 or higher

⚡ Quick Start

Install the tool:
```
npm install -g playwright-genie-ai
```
Install Playwright (if not already installed):
```
npx playwright install
```

Set up your API key:

echo "OPENAI_API_KEY=your_api_key_here" > .env

Generate and run your first test:

playwright-genie-ai \
  --url "https://www.saucedemo.com/" \
  --scenario "Login with standard_user and secret_sauce, then add first item to cart" \
  --run

⚙️ Setup

Install Playwright (if not already installed):
```
npx playwright install
```
Set up environment variables:
```
cp .env.example .env
```
Edit .env and add your OpenAI API key:
```
OPENAI_API_KEY=your_openai_api_key_here
```
For Ollama (local AI), ensure Ollama is running on localhost:11434 or set:
```
OLLAMA_BASE_URL=http://localhost:11434
```

🎯 Usage

Basic Usage

playwright-genie-ai --url "https://example.com" --scenario "Login with valid credentials"

Advanced Usage

playwright-genie-ai \
  --url "https://www.saucedemo.com/" \
  --scenario "Login as standard_user with password secret_sauce, sort by Name (Z to A), and add the first item to cart" \
  --model gpt-4 \
  --output "saucedemo-login-test" \
  --run

CLI Options

Option	Description	Required	Default
`--url <url>`	Target website URL	✅	-
`--scenario <text>`	Natural language test description	✅	-
`--model <model>`	AI model (`gpt-4`, `gpt-3.5`, `ollama`)	❌	`gpt-4`
`--output <filename>`	Custom output filename (without extension)	❌	Timestamped
`--run`	Run the test immediately after generation	❌	`false`

🧪 Example Test Cases

playwright-genie-ai \
  --url "https://www.saucedemo.com/" \
  --scenario "Login as standard_user with password secret_sauce, sort by Name (Z to A), and add the first item to cart" \
  --run

Form Submission

playwright-genie-ai \
  --url "https://httpbin.org/forms/post" \
  --scenario "Fill out the form with name 'John Doe', email 'john@example.com', and submit it"

playwright-genie-ai \
  --url "https://playwright.dev" \
  --scenario "Navigate to the docs section and search for 'selectors'"

🤖 Supported AI Models

OpenAI Models

GPT-4: Most capable, best for complex scenarios
GPT-3.5: Faster and cheaper, good for simple tests

Local Models (Ollama)

CodeLlama: Free local AI model for code generation
No API key required, runs entirely offline

📁 Generated Files

Tests are saved in the ./tests/ directory with the following naming:

Custom name: <your-name>.spec.ts
Auto-generated: generated-test-YYYY-MM-DDTHH-mm-ss.spec.ts

🔧 Development

Building from Source

git clone https://github.com/sxtyxmm/playwright-genie-ai.git
cd playwright-genie-ai
npm install
npm run build

Development Mode

npm run dev -- --url "https://example.com" --scenario "Your test scenario"

Project Structure

playwright-genie-ai/
├── src/
│   ├── index.ts        # CLI entry point
│   ├── generator.ts    # AI model integration
│   ├── fileWriter.ts   # Test file creation
│   └── runner.ts       # Test execution
├── tests/              # Generated test files
├── dist/               # Compiled JavaScript
├── package.json
├── tsconfig.json
└── README.md

🔍 How It Works

Input Parsing: CLI accepts URL and scenario description
AI Generation: Sends structured prompt to selected AI model
Code Generation: AI creates complete Playwright test code
File Writing: Saves test to /tests directory
Optional Execution: Runs test with Playwright if --run flag is used

🛠️ Troubleshooting

Common Issues

Missing OpenAI API Key

Error: OPENAI_API_KEY environment variable is required

Solution: Add your API key to .env file

Playwright Not Installed

Playwright is not installed. Please run: npx playwright install

Solution: Run npx playwright install

Generated Test Fails

Check if the website structure has changed
Try a more specific scenario description
Use --model gpt-4 for better accuracy

Debug Mode

Add DEBUG=1 to see detailed logs:

DEBUG=1 playwright-genie-ai --url "..." --scenario "..."

🤝 Contributing

Fork the repository
Create a feature branch: git checkout -b feature/amazing-feature
Commit changes: git commit -m 'Add amazing feature'
Push to branch: git push origin feature/amazing-feature
Open a Pull Request

📝 License

MIT License - see LICENSE file for details.

🔮 Future Features

Auto-detect selectors using page inspection
Support for multi-step test scenarios
Integration with ZeroStep's ai() helper
Test maintenance and updating capabilities
Visual regression testing support
Custom AI model endpoints

📞 Support

📊 NPM Stats

Made with ❤️ by the Playwright Genie AI team