running-e2e-tests

name	running-e2e-tests
description	Execute end-to-end tests covering full user workflows across frontend and backend. Use when performing specialized testing. Trigger with phrases like "run end-to-end tests", "test user flows", or "execute E2E suite".
allowed-tools	Read, Write, Edit, Grep, Glob, Bash(test:e2e-*)
version	1.21.0
author	Jeremy Longshore <[e-mail]>
license	MIT
tags	["testing","workflow","e2e-tests"]
compatibility	Designed for Claude Code, also compatible with Codex and OpenClaw

E2E Test Framework

Current State

!cat package.json 2>/dev/null | grep -oE 'playwright|cypress|selenium' || echo 'No E2E framework detected'

Overview

Execute end-to-end tests that simulate real user workflows across the full application stack -- browser interactions, API calls, database operations, and third-party integrations. Supports Playwright (recommended), Cypress, Selenium, and Puppeteer.

Prerequisites

• E2E testing framework installed (Playwright, Cypress, or Selenium WebDriver)
• Application running in a test environment with seeded test data
• Browser binaries installed (npx playwright install or Cypress binary)
• Test user accounts created with known credentials
• Environment variables configured for base URL, API keys, and test credentials

Instructions

1. Identify critical user journeys to cover:
   • User registration and login flow.
   • Primary feature workflow (e.g., create item, edit, delete).
   • Search and filtering functionality.
   • Checkout or payment flow (if applicable).
   • Error handling (404 pages, form validation, session expiry).
2. Create page object models (POM) for reusable page interactions:
   • One class per page or major component.
   • Encapsulate locators, actions (click, fill, select), and assertions.
   • Use data-testid attributes as primary selectors for stability.
3. Write E2E test files organized by user journey:
   • Each test file covers one complete workflow.
   • Use beforeEach to navigate to the starting page and reset state.
   • Use afterEach to capture screenshots on failure.
   • Keep tests independent -- no test should depend on another test's output.
4. Handle authentication efficiently:
   • Store authenticated session state to a file (storageState in Playwright).
   • Reuse session across tests that require login.
   • Create a separate auth setup fixture that runs once per worker.
5. Configure multi-browser and responsive testing:
   • Run tests on Chromium, Firefox, and WebKit.
   • Test at mobile (375px), tablet (768px), and desktop (1280px) viewports.
   • Use Playwright projects to define browser/viewport combinations.
6. Add retry and stability mechanisms:
   • Use expect with auto-waiting locators (Playwright) instead of explicit waits.