More screenshots

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIDeepSeekDeepSeekMuseMuse+22 more

    Playwright: Test Websites the Way Users Use Them

    1

    Drive real browsers with Playwright — user-first locators, smoke tests, form and login flows, responsive checks, link crawls, and artifacts that prove it worked.

    $5

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    Playwright: Test Websites the Way Users Use Them

    Playwright: Test Websites the Way Users Use Them

    Example session with this skill installed

    Verify our staging store's checkout flow end to end — desktop and mobile — and prove it with assertions and screenshots, not a claim.

    • Read your context and instructions
    • Compiled the playwright

    Setup once: npm install, npx playwright install chromium, scripts in a scratch directory, screenshots, traces, and videos in a timestamped artifacts folder. Login isolated in a helper that logs in via environment credentials and returns an authenticated context — reused across tests instead of logging in per test, and credentials never hardcoded in scripts. The test adds an item to cart and asserts the cart count is 1, goes to checkout, fills shipping with getByLabel fields, submits with a test card, and asserts the confirmation page shows an order number matching the expected pattern — asserting resulting state, not just 'no error'. Screenshots at each step, trace recorded. Same script at desktop 1440px and mobile 390px. Desktop passes. Mobile fails: the 'Place order' button is overlapped by a sticky banner. That is a real bug found, not a test bug — exactly what the skill is for. The report states what the script did, what it asserted, and where the artifacts are. The flow is not claimed to work — the assertion output backs it, and the mobile failure is filed with its screenshot and trace. The mobile failure is the point: a test that only passes on desktop never met the user. The bug is filed with its screenshot and trace, and the rerun will prove the fix.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    Automated browsers exist to answer one question: does the app behave for a real user? This skill writes scripts that act like users — click what users see, assert what users perceive — so the tests survive refactors instead of breaking on every CSS change. It covers setup, the user-first locator rule (roles and accessible names before CSS and XPath), script patterns for smoke tests, form flows, login flows, responsive checks, and link crawls, waiting without flakiness, artifacts that prove the run worked, and CI discipline.

    Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse All Claude Skills.

    It needs Node.js 20+ and a Chromium install on first setup.
    A test that needs two retries is a flaky test — fix it, do not configure around it; CI retries failures once.
    CSS and XPath are the last-resort locators; if you keep reaching for them, the app probably needs test IDs.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before buying a skill.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    Fresh listing

    Recently published to Agensi

    Money-back guarantee

    Doesn't do what its page promises? You get your money back.

    Trust & safety

    Security scan

    Security scan passed 3 days ago

    Software
    Node.jsPlaywright

    Permissions

    Terminal / Shell
    Browser
    Listed3 days ago

    What's inside

    Frequently Asked Questions

    No. Assert on state, not time — expect(locator).toBeVisible() beats waitForTimeout(3000) in every way. Auto-waiting is your friend; manual sleeps on top of it are the flake.

    No. Isolate login in a helper, save the session, and reuse it across tests — faster and less flaky. Never hardcode credentials; read them from environment variables.

    Yes — it is the last-resort locator. Prefer role plus accessible name, then labels, then text, then test IDs. If XPath is all you have, the app probably needs test IDs added.