cypress-test-bootstrap

Cypress Test Bootstrap

Purpose

Scaffold a production-ready Cypress project or generate structured test files following best practices — organized for maintainability and QA team reuse.

---

Input Accepted

• Request to init a new Cypress project from scratch
• Feature name or module to add tests for
• User Story or Gherkin scenarios to automate
• API endpoint to test via cy.request()
• Existing test suite to extend

---

Project Structure (Full Bootstrap)

[project-name]/
├── cypress.config.js
├── package.json
├── .env.example
├── .gitignore
└── cypress/
    ├── e2e/                        ← test specs by feature/module
    │   └── [module]/
    │       └── [feature].cy.js
    ├── fixtures/                   ← test data (JSON files)
    │   ├── users.json
    │   └── transactions.json
    ├── support/
    │   ├── e2e.js                  ← global setup / imports
    │   ├── commands.js             ← custom reusable commands
    │   └── api-helpers.js          ← API utility functions
    └── pages/                      ← Page Object Model
        └── [FeatureName]Page.js

---

Generation Process

Full Project Bootstrap — generates:

1. cypress.config.js — config with env variables
2. package.json — dependencies and run scripts
3. .env.example — environment variable template
4. .gitignore — excludes artifacts, credentials, node_modules
5. cypress/support/commands.js — login + apiRequest custom commands
6. cypress/support/e2e.js — global imports
7. cypress/fixtures/users.json — test data structure
8. Sample transfer spec file ready to adapt to the target feature

Single Spec File — generates:

• describe block for the feature
• beforeEach with API login for authenticated flows
• context('Happy Path') / context('Negative Cases') / context('Boundary Cases')
• it blocks with clear descriptive names
• Assertions with .should() — never cy.wait(ms)
• Custom commands for repeated actions

---

Key Conventions

✅ For authenticated features, login via API in beforeEach (cy.request) — not via UI
✅ Use data-testid selectors exclusively
✅ Group tests: context('Happy Path') / context('Negative Cases')
✅ Store all credentials and URLs in environment variables
✅ Use .should() assertions — never cy.wait(ms)
✅ Each test is fully independent — no shared state
✅ Clean up test data after tests when needed