An AI QA engineer for the customer flows you need to trust

Zerocheck is an AI QA engineer that tests your app end to end. Verify critical flows and find bugs with a prompt.

Why we built it

Coding agents write more code than most teams can check by hand. Unit tests and code review cover a lot, but someone still needs to check that customers can sign in, find a product, update a setting or complete a purchase. That usually means someone clicking through staging, or a browser suite someone has to keep up.

Start with the flows you already check before shipping, or the one that broke last time. Give each one a specific expected outcome, and add more one flow at a time.

How it works

1. We set up your first tests with you. We provision a dedicated instance for your team and set up your first 3–5 tests on your own app. You keep adding tests from there.
2. Describe a flow, or paste a checklist. In the web app, say what a customer should do and what should happen, including any setup or test data it needs.
3. See what happened. Zerocheck drafts readable steps and runs them in a hosted browser right away. Read the result, edit the steps and answer anything it needs to know.
4. Save it and run it. Keep it in Your tests, run it against a chosen environment, and edit it when your product's expected behavior changes.
5. Run it locally or in CI. Download a test as YAML and run it with the CLI, on your machine or in a CI job. Make it a required check, and a blocking failure stops the merge. A download is an independent copy, so web and repository edits don't sync.

If your team only uses the web app, you need no repository and no CLI installation. The CLI is useful for localhost, private networks your machine can reach, and existing CI jobs.

"Can I see what actually happened?"

Open a run to see the steps that ran, the result of each one, before-and-after screenshots and error details. Console warnings and errors help you investigate. A recording is available when the test uses no configured credentials.

A failed expected outcome, a missing credential and an interaction the agent could not complete are different results, and Zerocheck shows you which one you got. That helps you decide whether to fix your app, your setup or the test description. The cause of a bug is still yours to work out.

"How much of this relies on AI?"

AI drafts tests and uses the page's current state to find controls and judge outcomes written in plain English. Checks for exact text, URLs and network requests use direct browser evidence. For important amounts, labels and destinations, write a specific check your reviewer can read.

Your team defines what success means, and Zerocheck never changes an expected outcome to make a test pass.

"What happens when our UI changes?"

You spend less time maintaining tests. Zerocheck can recover from layout and selector changes, and shows you what changed in the run results. If the expected behavior changes, the test fails so you can review it. If a transient AI-service error is retried and the test then passes, the result is labelled flaky. Failed outcomes are never retried into a pass. Results, retries and healing has the details.

Where Zerocheck fits

Zerocheck is for teams shipping a web product who have customer flows to verify and an engineer who can define expected outcomes and test accounts. It works well as a starting point when the release checklist still lives in a document or in someone's head.

Tests run in Chromium. Authenticated flows need configured login steps and credentials. CLI runs depend on your instance, which makes the model calls and receives the selected tests and what the browser shows. Read the supported flows and data handling before choosing your first flows.

What it costs

Zerocheck is US$500 a month, with US$500 in credits at sign up that cover your first 30 days and up to 1,000 test runs. A dedicated instance for your team, with hands-on setup of your first tests. Run tests from the web app or CLI, with the screenshots and errors you need to investigate a failure. The monthly service includes about 1,000 test runs a month, with more by agreement. It's month to month, and you can cancel at any time.

Book a setup call · Read the pricing · Create your first test