CyGen - AI-Powered Cypress Test Generator

[Under Construction]

CyGen is an intelligent test generator that automatically creates Cypress tests for your JavaScript/TypeScript API endpoints. It uses AI agents to generate comprehensive test cases.

Features

• Automatic test generation for JavaScript/TypeScript API endpoints
• AI-powered test case generation using agents
• Comprehensive test scenarios including:
  • Happy path tests
  • Negative test cases
  • Edge cases
  • Error handling
• Real-time file watching
• Customizable output directory
• Support for modern JavaScript features

Installation

Global Installation (Recommended)

# Install globally
npm install -g cygen

# Verify installation
cygen --help

Local Installation

# Install in your project
npm install cygen

# Run using npx
npx cygen --help

Usage

Command Line Interface

CyGen provides two main commands:

1. Watch Mode - Continuously watch for file changes:

# Basic usage with Ollama
cygen watch --use-ai

# With OpenAI
cygen watch --use-ai --llm openai --model gpt-4 --api-key YOUR_OPENAI_KEY

# With web search enabled
cygen watch --use-ai --web-search --llm ollama --model llama3.1

2. Test Mode - Generate and optionally run tests for specific files:

# Generate tests with OpenAI
cygen test --files ./src/api/users.js --use-ai --llm openai --model gpt-3.5-turbo --api-key YOUR_OPENAI_KEY

# With custom Ollama server
cygen test --files ./src/api/*.js --use-ai --llm ollama --model mistral --base-url http://localhost:11434

# Generate and run tests
cygen test --files ./src/api/*.js --use-ai --web-search

CLI Options

Watch Command Options

| Option | Alias | Description | Default | |--------|-------|-------------|---------| | --watch-dir | -w | Directory to watch for file changes | Current directory | | --output-dir | -o | Directory to output generated tests | cypress/integration/generated | | --use-ai | | Enable AI-powered test generation | false | | --web-search | | Enable web search for test generation | false | | --llm | | LLM provider (ollama or openai) | ollama | | --model | | Model name for the selected LLM | llama3.1 | | --api-key | | API key for the LLM provider | | | --base-url | | Base URL for the LLM service | http://localhost:11434 |

Test Command Options

| Option | Alias | Description | Default | |--------|-------|-------------|---------| | --files | -f | Files to generate tests for | (required) | | --output-dir | -o | Directory to output generated tests | cypress/integration/generated | | --use-ai | | Enable AI-powered test generation | false | | --web-search | | Enable web search for test generation | false | | --llm | | LLM provider (ollama or openai) | ollama | | --model | | Model name for the selected LLM | llama3.1 | | --api-key | | API key for the LLM provider | | | --base-url | | Base URL for the LLM service | http://localhost:11434 | | --run | | Run the generated tests with Cypress | false |

Programmatic Usage

You can also use CyGen programmatically in your code:

const { CyGen } = require('cygen');

// Using OpenAI
const cygen = new CyGen({
    useAI: true,
    aiOptions: {
        llm: 'openai',
        model: 'gpt-4',
        apiKey: 'your-openai-key',
        enableWebSearch: true
    }
});

// Using Ollama
const cygen = new CyGen({
    useAI: true,
    aiOptions: {
        llm: 'ollama',
        model: 'mistral',
        baseUrl: 'http://localhost:11434'
    }
});

// Generate tests
await cygen.generateTestsForFile('./src/api/users.js');

Configuration

The following options are available:

• watchDir: Directory to watch for changes (default: current working directory)
• outputDir: Directory where test files will be generated (default: './cypress/integration/generated')
• useAI: Enable AI-powered test generation (default: false)
• aiOptions: AI configuration options
  • llm: LLM provider (ollama or openai)
  • model: Model name for the selected LLM
  • apiKey: API key for the LLM provider
  • baseUrl: Base URL for the LLM service
  • enableWebSearch: Enable web search for test generation

Supported File Types

• JavaScript/TypeScript API files
• Swagger/OpenAPI specification files (JSON/YAML)

Generated Tests

Tests are generated with comprehensive coverage including:

1. Happy Path Tests

   • Successful API calls
   • Valid request/response handling
   • Expected data validation

2. Negative Tests

   • Invalid input handling
   • Error response validation
   • Authentication/Authorization failures
   • Rate limiting scenarios

3. Edge Cases

   • Boundary value testing
   • Empty/null input handling
   • Large payload handling
   • Timeout scenarios

Example generated test:

describe('GET /api/users', () => {
    beforeEach(() => {
        cy.intercept('GET', '/api/users').as('apiRequest');
    });

    it('should return a list of users', () => {
        cy.request({
            method: 'GET',
            url: '/api/users',
            failOnStatusCode: false
        }).then((response) => {
            expect(response.status).to.equal(200);
            expect(response.body).to.be.an('array');
            expect(response.body[0]).to.have.property('id');
            expect(response.body[0]).to.have.property('name');
        });
    });

    it('should handle invalid requests', () => {
        cy.request({
            method: 'GET',
            url: '/api/users/invalid',
            failOnStatusCode: false
        }).then((response) => {
            expect(response.status).to.equal(404);
            expect(response.body).to.have.property('error');
        });
    });
});

Supported LLMs

Ollama

• Default models: llama3.1, mistral, llama2
• Requires local Ollama server (default: http://localhost:11434)

OpenAI

• Models: gpt-4, gpt-3.5-turbo
• Requires API key