Automateyourtests.
No code.No excuses.

No. You build tests by dragging nodes onto a canvas and connecting them — navigate, click, fill, assert. If you can describe the steps a user takes, you can build the test. Flow Testing is built for manual QA testers first, not developers.

Playwright, the same engine used by engineering teams at scale. Your visual flow is compiled into a real Playwright run — so you get the reliability of a mature test runner without writing the code.

Chromium, Firefox and WebKit, which covers Chrome, Edge, Firefox and Safari engines. Cross-browser runs are included on every plan, Free included.

No, and this is deliberate. You can export any flow as JSON or as a real Playwright test file at any time, on any plan. If you outgrow Flow Testing, you walk away with working code rather than starting over.

Yes. Store credentials in the secrets vault and reference them from a node, so they never sit in plain text in your flow. On Pro you can also save a login sequence as a reusable sub-flow and drop it into every test instead of rebuilding it each time.

That is what the healer agent is for. When a selector stops matching, it proposes a fix based on the current page instead of leaving you with a red run and no explanation. You review the change before it sticks.

One execution of one flow, whether you started it yourself or a schedule started it. Building, editing and saving flows are always free and unlimited — you are only ever charged for execution.

Further runs are blocked until the period resets or you upgrade. Nothing is deleted and nothing is charged automatically: your flows, history and artifacts stay exactly where they are.

A run that fails because your test or your application failed does count, since the work was performed. A run that fails because of a defect on our side does not.

Lemon Squeezy is our merchant of record — they process the payment and handle VAT and sales tax, so that is the name that appears on your card statement. Your invoices and receipts come from them too.

Cancel any time from billing settings; you keep Pro until the end of the period you already paid for. If you are unhappy, email us within 14 days of a charge and we refund it in full, no justification needed. The full terms are on the Terms page.

Into the secrets vault, encrypted at rest and referenced from your flow as a {{SECRET:name}} placeholder. Once saved the plaintext is never returned by our API — only the name and creation date are readable. That keeps credentials out of your flow definition, out of exported files and out of our database in readable form. It does not keep them out of run artifacts — see the next question, which you should read before storing anything sensitive.

Yes, and we would rather tell you than have you find out. To type a credential into a real login form, the value has to be decrypted first — so it appears in that run's artifacts: in the recorded step parameters, in the page snapshot, and in the login request body. A password field renders as dots on screen, but the characters are still in the underlying recording, and anything in a plain text field is visible outright. Anyone who can open that run can read them. This is inherent to driving a real browser, not specific to us — no tool that fills a real login form can avoid it entirely. So: use dedicated, low-privilege test accounts, never real user or admin credentials. Artifacts are deleted after 30 days and you can delete a run sooner.

You can, but think carefully first. Playwright records traces, screenshots and network logs of every run, so testing production means real customer data — and the credentials the test signed in with — end up in those artifacts. We recommend staging or seeded data and a throwaway test account wherever possible. The Privacy Policy explains what this means for your own obligations.

Traces, screenshots, video and logs are kept for 30 days and then deleted. You can delete a run sooner yourself at any time.

No. Your flows, secrets and run artifacts are yours, we claim no ownership of them, and we do not read them except where strictly needed to run or repair the service. We do not sell data and run no ad tracking.

Three things: the planner turns a description of a user journey into a draft flow, the generator builds individual steps and selectors, and the healer repairs steps that broke when your UI changed. They are on the Pro plan.

Yes. Bring your own key for Anthropic or OpenAI, or point at a custom endpoint, and AI requests run against your own account with that provider. Keys are stored encrypted.

No. We do not use your flows, test data or run artifacts to train models — ours or anyone else's.

Review it. AI output is a suggestion, not a verified result: it can pick the wrong element, or heal a step into something that passes while no longer testing what you intended. A green AI-generated test is not by itself proof your app is correct.