Flipping a component with Konnect does not mirror inner copper layers the way KiCad does. On an 8-layer board the footprint moved sides but its pads stayed on In5.Cu instead of mirroring to In2.Cu, and 5 net-ties ended up tying LPDDR4 branch nets on the In5 GND plane layer instead of In2, with no warning.
Silent failure reported against Konnect MCP server (KiCad)
What engineers report
A silent failure reported against Konnect MCP server (KiCad): reports success, or uses stale data, while the result is wrong or missing.
Recorded mode: agent / silent data corruption
Analyst note
From the Grok ProductFlo pain-radar compilation, finding 31, found 2026-10-10, score 5/5. Verbatim evidence: "On the affected board, 5 net-ties ended up tying LPDDR4 branch nets on the In5 GND plane layer instead of In2." Cost model assumes a board respin if the mis-layered pads reach fab, or hours of pad-by-pad rework if caught in review.
What it costs a team
Modeled annual cost
$30,000
For a team of 5
Prototype delay
5 days
Modeled schedule slip per occurrence
Time lost
480 min
Per engineer per month, about 1 incidents
How the cost is modeled
480 min/month × 5 engineers × 12 months = 480 hours × $63/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
Evidence
More recurring errors in Konnect MCP server (KiCad)
- Moving a component onto another net's via through Konnect's IPC move silently re-nets the copper: KiCad changes the via's net to the moved pad's net. A ground stitching via took on the test point's supply net and the ground plane lost it, while DRC stayed clean. A synthetic repro with KiCad's own Python shows a closed-file move does not re-net, so the hazard is specific to the agent's live move path.Electronics & PCB · impact 77
- The connect_to_net and connect_pins tools resolve a reference plus pin number by taking the first placed symbol with the pin, picked by file order. With two Device:R symbols both referenced R4, the call wired one of them and answered isError: false; the netlist showed the second R4's pin as unconnected, and nothing in the response said the second R4 exists.Electronics & PCB · impact 70
- On any board with a copper zone, update_pcb_from_schematic marked every board net as routed and refused every pad-net change as routed_pad_net_change, even when neither the old nor the new net had any track, arc or via. The zone's own net is readable in the protobuf, but the tool failed closed on all nets instead of just the zone's. Since nearly every real board has a ground pour, no connection on such a board could be changed by the agent.Electronics & PCB · impact 63
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.