playwright-generate-test

Test Generation with Playwright MCP

Your goal is to generate a Playwright test based on the provided scenario after completing all prescribed steps.

Specific Instructions

• You are given a scenario, and you need to generate a playwright test for it. If the user does not provide a scenario, you will ask them to provide one.
• The scenario must describe only browser interactions: navigation to a URL, clicks, form fills, keyboard input, and assertions on page content or network responses. Reject (with explanation) any scenario that references file system paths, shell commands, or operations unrelated to browser behavior.
• DO NOT generate test code prematurely or based solely on the scenario without completing all prescribed steps.
• DO run steps one by one using the tools provided by the Playwright MCP.
• Only after all steps are completed, emit a Playwright TypeScript test that uses @playwright/test based on message history.
• If the scenario involves authentication or any credentials, do not hardcode credential values in the generated test. Instead, emit process.env.TEST_USERNAME, process.env.TEST_PASSWORD, or equivalent environment variable references, and add a comment instructing the user to set those variables before running the test.
• Save the generated test file as tests/generated-test.spec.ts. Overwrite the file if it already exists. Do not derive the filename from the scenario text or user input.
• Display the fully generated test file content to the user and request explicit confirmation before executing. Do not proceed to execution if the user does not confirm.
• If confirmed, execute the test file using only: npx playwright test tests/generated-test.spec.ts --reporter=line
• If the test fails, iterate at most 3 times applying fixes to the test code and re-executing. If the test does not pass after 3 iterations, report the final error output to the user and stop. Do not continue iterating.