playwright-skill

Playwright Browser Automation

Overview

Use this skill to write and run focused Playwright scripts for testing or automating websites. Keep the main flow independent of any specific host or launcher. Choose the execution surface that already exists in the current environment.

When to Use

Use this skill when the user wants to:

• test a web page or flow in a real browser
• take screenshots or verify responsive layouts
• automate form filling or navigation
• validate login, checkout, or dashboard behavior
• inspect broken links or page content with scripted checks

Core Workflow

1. Identify the target URL.
2. If the target is local, check whether a development server is already running.
3. Write a temporary Playwright script outside the repository or in another disposable location.
4. Parameterize the target URL and any secrets instead of hardcoding them.
5. Run the script using the available Playwright runtime in the current environment.
6. Collect artifacts such as screenshots, logs, and assertion results.
7. Clean up temporary files when they are no longer needed.

Use a visible browser by default unless the user explicitly asks for headless execution.

Inputs / Assumptions

• A reachable URL is required before the script runs.
• Localhost testing should confirm the correct port first.
• Credentials and secrets should come from environment variables or explicit user input, never from hardcoded literals.
• Temporary scripts should not clutter the project unless the user asks to keep them.

Examples

Basic page check

const { chromium } = require('playwright');

const TARGET_URL = process.env.TARGET_URL ?? 'http://localhost:3000';

(async () => {
  const browser = await chromium.launch({ headless: false });
  const page = await browser.newPage();

  await page.goto(TARGET_URL, { waitUntil: 'domcontentloaded' });
  console.log('Title:', await page.title());
  await page.screenshot({ path: '/tmp/page-check.png', fullPage: true });

  await browser.close();
})();