PlaywrightTestver for Playwright
AI that writes, runs, and repairs Playwright tests — in your repo, in your style.
Playwright gives you a great API and honest tooling. What it does not give you is someone to write the specs, read the traces when they fail, and fix the locators after a redesign. That is the part Testver takes on.
Point Testver at your existing Playwright project and it reads the framework, language and structure you already use. Generated specs land as plain files in your repo — same fixtures, same config, same runner. Nothing about how you execute tests changes.
Languages supported for Playwright: TypeScript · JavaScript · Python · Java · C#
What a Playwright suite costs to keep
- Every spec is still handwritten—The API is ergonomic, but coverage still grows one manually typed test at a time.
- Traces are rich — and manual to read—A failing run gives you a beautiful trace that a human still has to open, scrub through, and interpret.
- Locators drift with the UI—A renamed test id or restructured component turns a green suite red, and someone has to find every call site.
The same suite, maintained for you
Record once, get idiomatic code
Click through the flow and the AI Record Agent writes a spec in your project’s conventions — getByRole and getByTestId where they fit, page objects if that is how your repo is organised.
Auto-Heal reads the trace you already produce
When a locator breaks, Testver pulls the DOM from the failing run’s trace, proposes the corrected locator, and re-runs the affected tests before keeping the change.
Ask in plain English, in context
The assistant has your repository’s context, so "add a test for the expired-coupon path" produces something that fits the suite instead of a generic sample.
Runs local or on a cloud grid
Execute against your local browsers or on BrowserStack, LambdaTest and Sauce Labs, with results collected the same way either way.
Everything around the runner
Scheduling, run history, trend reporting, test management and CI gating — the layer Playwright deliberately leaves to you.
What lands in your repository
Generated from a recorded session — plain Playwright, no Testver imports. Delete Testver tomorrow and this test still runs — there is no Testver runtime, adapter or import inside your project.
TypeScripttest('checkout applies the promo code', async ({ page }) => {
await page.goto('/cart');
await page.getByTestId('promo-input').fill('SAVE20');
await page.getByRole('button', { name: 'Apply' }).click();
await expect(page.getByTestId('order-total')).toHaveText('$79.20');
});Point Testver at your Playwright project
It detects the framework, language and structure on its own. No migration, no config file to write, no change to how you run tests.
Install it
One global npm package. Node 18 or newer is the only prerequisite.
npm install -g @testver/testverGo to your test project
Testver runs from inside the framework project you want to test, so change into that folder first.
cd path/to/your-test-projectStart the dashboard
Opens at localhost:3700. Testver detects the framework, language and structure of the project you started it in and lists your tests — no configuration to write.
testver serveTypically under five minutes, start to first generated test.
Full installation guide — prerequisites, ports, running as a servicePlaywright questions, answered
- Do I have to rewrite my existing Playwright tests?
- No. Testver reads your existing project and writes new specs into it in the same style. Your config, fixtures and runner are untouched, and every generated test is plain Playwright code.
- Which languages does it support for Playwright?
- TypeScript and JavaScript, plus Playwright’s Python, Java and .NET bindings. The generator follows your project’s own conventions rather than imposing a fixed style.
- How does Auto-Heal fix a broken Playwright locator?
- It reads the DOM from the failing run’s trace, proposes the corrected locator, then re-runs every affected test before the change is kept — and you can undo it afterwards.
More in the full FAQ or the documentation.
Running more than one framework?
One Testver instance handles all of them — different projects can use different frameworks, languages and AI models.
AppiumSee the full feature set, the comparison with other approaches, or try a live demo repo.
Keep your framework. Add AI and a premium toolset.
Set up in minutes. Self-host on your own hardware. Describe a test in plain English and get working code in your own framework — in a fraction of the time.