Live checks in Altium 26.11 found that a refused or rejected obj_modify still read as success, and dry runs did not say what they would apply. The fix reports modified:false and stops the false 'Written to Altium.' message.
Recurring agent failure issue in eda-agent (MOREmbedded)
What engineers report
A recurring agent failure problem in eda-agent (MOREmbedded) that mostly hits the electrical engineer.
Failure pattern: agent / 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 eda-agent (MOREmbedded).
Evidence
More recurring errors in eda-agent (MOREmbedded)
- The Altium/EasyEDA agent's request ids put the session tag first and the trace logged only the first 8 characters, so perf joined on the session instead of the call. On a real workspace 1244 trace lines came back as 129 calls with nothing saying two thirds were gone.ECAD · impact 74
- Altium MCP bridge: every status row showed the session id instead of the call id, so grepping the trace log returned every call; also notes the deployed Altium scripts on the machine are ahead of the branch (PCB.pas 5929 lines and others), so edits risk overwriting live code.ECAD · impact 66
- Merging 23 upstream commits into an Altium agent fork, 655 to 758 tools; 18 conflicts were settled by 'the side that measured it won'. On impedance reference layers they had it backwards at first; upstream had rerun reads on a board with profiles and found the layers unreadable.ECAD · impact 66
- The eda-agent doctor reported a critical FAIL of no process named x2.exe found while Altium Designer 26.11 was on screen and answering bridge commands, and the FAIL returned before the round-trip ping ran, so the contradicting signal never got a say.ECAD · impact 66
- The proj_run_erc check stopped the Altium agent bridge on Altium Designer 26.11.1.5 with Undeclared identifier from two variables a commit added without declaring. A new lint for undeclared reads also found that FindPadNetByName was dropped by the 2026-10-10 merge while PCB_PlaceVia, PCB_PlaceTrack and PCB_RebindCopperToPadNets kept calling it, so the agent's ERC and its net-aware copper placement were broken for a month inside Altium's scripting layer.ECAD · impact 66
- The tool meant to diagnose the Altium bridge reported Altium as down, with no version and no round trip, on a machine where it was running and answering. It trusted a process-table scan over the bridge; now it pings first and reports both.ECAD · 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.