Saved false text turned on real inner doors in manufacturing output: sheet-metal cabinet code used bool on the multi_door_enabled field and pre-coerced Door Layout rows with float, so saved legacy false enabled real physical inner doors and True in an unselected grid cell became a valid-looking 1-mm physical cell. Python truthiness reached CUTTING, BEND and DXF output.
Recurring agent failure issue in whd (looaeedr)
What engineers report
A recurring agent failure problem in whd (looaeedr) that mostly hits the mechanical designer.
Failure pattern: validation / failure
Analyst note
From the Grok pain-radar digests, Nov 8-9 2026 (Nov 8 8:41 AM ET through Nov 9 6:41 AM ET digests plus the Haitch editions). Verified live 2026-11-09: issue/PR read; title and state confirmed.
What it costs a team
Modeled annual cost
$30,000
For a team of 5
Prototype delay
3 days
Modeled schedule slip per occurrence
Time lost
240 min
Per engineer per month, about 1 incidents
How the cost is modeled
240 min/month × 5 engineers × 12 months = 240 hours × $125/hour = $30,000
A modeled estimate with medium confidence. Swap in your own team size and loaded rate to size it for your team.
Engineering tools that address it
No engineering tool profile lists this error yet. Browse tools that connect to whd (looaeedr).
Evidence
More recurring errors in whd (looaeedr)
- The manufacturing adapter silently swallowed profile readback and workspace source failures, letting incomplete snapshot or default inputs enter Engine manufacturing resolution. In the companion GUI fix, default width, height, depth and thickness were substituted when physical control reads failed and then saved.MCAD · impact 74
- Follow-up to last hour's whd findings. "The main GUI still independently calculated marking-hole X/count/pitch/Y while the Engine's real IndicatorBox MARKING scene generated the same pattern." Other PRs remove a GUI copy of part-identity logic and stop the engine from replacing corrupt inputs with a default group "or fabricating 1mm physical Door/BasePlate sizes for invalid geometry." Agent-written PRs, each starting with a SOURCE_READ_EVIDENCE block listing what was read first, which is its own interesting habit.MCAD · impact 66
- Follow-up to last hour's whd findings. "The main GUI still independently calculated marking-hole X/count/pitch/Y while the Engine's real IndicatorBox MARKING scene generated the same pattern." Other PRs remove a GUI copy of part-identity logic and stop the engine from replacing corrupt inputs with a default group "or fabricating 1mm physical Door/BasePlate sizes for invalid geometry." Agent-written PRs, each starting with a SOURCE_READ_EVIDENCE block listing what was read first, which is its own interesting habit.MCAD · impact 66
- Follow-up to last hour's whd findings. "The main GUI still independently calculated marking-hole X/count/pitch/Y while the Engine's real IndicatorBox MARKING scene generated the same pattern." Other PRs remove a GUI copy of part-identity logic and stop the engine from replacing corrupt inputs with a default group "or fabricating 1mm physical Door/BasePlate sizes for invalid geometry." Agent-written PRs, each starting with a SOURCE_READ_EVIDENCE block listing what was read first, which is its own interesting habit.MCAD · impact 66
- Follow-up to last hour's whd findings. "The main GUI still independently calculated marking-hole X/count/pitch/Y while the Engine's real IndicatorBox MARKING scene generated the same pattern." Other PRs remove a GUI copy of part-identity logic and stop the engine from replacing corrupt inputs with a default group "or fabricating 1mm physical Door/BasePlate sizes for invalid geometry." Agent-written PRs, each starting with a SOURCE_READ_EVIDENCE block listing what was read first, which is its own interesting habit.MCAD · impact 66
- The engine strictly validates door-cell indicator data, but the Fold Designer snapshot first ran a GUI repair that silently swapped invalid lamp counts and offsets for defaults. That could make bad saved manufacturing source look like valid holes in the DXF output. The fix rejects it before the snapshot.MCAD · impact 66
Sourced from public community reports. Costs and delays are modeled estimates, not measurements. Know a better source or a fix? Tell us at hello@productflo.io.