What problem does it solve? Writing integration and E2E tests usually means guessing selectors, hardcoding URLs, and booting a fresh environment for every run. This Skill removes that guesswork by exploring the running application first, reusing a shared test environment, and producing executable tests that match the repository's existing runner conventions. ## Core Features & Use Cases - Live-app exploration before authoring: Drives the configured browser provider (open, snapshot, interact, assert, screenshot, close) to observe real element roles, labels, and flows, so tests use only verified selectors. - Shared test environment reuse: Attaches to the validated test-env.json descriptor written by ot-prepare-test-env instead of booting a second instance, with credential references kept out of model context. - Repo-native runner preservation: Mirrors existing Playwright, Cypress, or custom test layouts and naming conventions, and generates paired POSIX/PowerShell launchers for agent-browser scenarios. - Artifact-based failure analysis: Classifies every failure as product regression, test issue, or environment problem, with a mandatory per-test evidence and ownership table. - Use Case: After implementing a company-creation flow, ask the agent to test it — it attaches to the running environment, walks the flow in a real browser, writes a Playwright test using observed getByRole locators, verifies it passes, and reports coverage. ## Quick Start Ask the agent to create and run an integration test for a specific feature, for example: "Write an integration test for the login rate-limiting flow and run the existing suite."
