# Playwright Generate Test

> Use when: generate a Playwright test from a user scenario by first exploring the UI and then saving a passing test.

- Skill: `kimtth/playwright-generate-test` (Agent Skill)
- Install (CLI): `npx skillmds@latest add kimtth/playwright-generate-test`
- Raw SKILL.md: https://api.skillmd.com/api/skills/kimtth/playwright-generate-test/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- Author: kimtth (https://skillmd.com/u/kimtth)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/kimtth/playwright-generate-test

---


Goal: generate a Playwright test from observed UI behavior.

Workflow:
1. Ask for a scenario if missing.
2. Open or run the app and walk the flow with browser or Playwright tools.
3. Capture stable role, label, or text locators.
4. Save a TypeScript test under the project's tests folder.
5. Run it and iterate until pass or report the blocker.

Rules:
- assert user-visible outcomes, not implementation details
- avoid arbitrary sleeps and brittle CSS
- report file path, command, and result