The desk is rehearsal. Your app is the next test.

Updated October 15, 2026 · Independent experiment

Knowing that a cookie survived is useful. Knowing that your own application accepted it is a different test. Keep the two observations separate.

Write down one application expectation

Choose a concrete action: reopen a draft, resume an onboarding step, recover a saved filter, or display an account after restoring state. Record the application URL, version, browser mode and expected result. Avoid a vague “session works” criterion; a logged-in header can hide an expired backend session.

First reproduce the storage mode on CarryLab. Then run the same kind of session transition in your own application using a dedicated test account. Observe a real application action that requires the expected state. Compare the failure boundary, not just the appearance of the page.

A fresh browser view

Open WebTraceCheck when a separate browser session would help inspect your own public page, fresh visit or onboarding entry point. Use a page you control and a non-sensitive test flow. Choose the browser capability actually offered; a remote Chromium session does not establish Safari, iOS or device-bound authentication behavior.

Record the visible result and relevant requests in that environment. WebTraceCheck does not import the desk's fixtures or automatically receive your observation file. A clean browser also cannot prove that all other origins are isolated or that your application's server-side session is valid.

An optional inbox for a staging sign-up

If your own test flow requires email, use CleanTempMail for a non-sensitive disposable test address. Complete one sign-up in a staging or otherwise authorized test environment, then exercise the intended session transition. Do not use a temporary inbox for recovery of an account you need to retain.

Whether a message arrived, whether its link was accepted, and whether a browser restored state are three separate observations. CarryLab does not send email, inspect your inbox or test deliverability. A blocked disposable address is an application-policy result, not evidence that your browser lost a cookie.

Keep a small evidence record

Note the exact initial application action and whether it completed.
Record the transition: same live context, new context with state, or reopened profile.
Perform the application's next meaningful action, rather than trusting a remembered label.
Attach only sanitized screenshots, logs or request metadata needed to explain that outcome.

Both linked products are related services promoted by this site. Links use referral tags to identify the placement. They are optional; all CarryLab experiments run without an account or either service.