kicad-studio-kit (oaslananka) errors

14 recurring problems engineers report in kicad-studio-kit (oaslananka), ranked by impact.

#28 by errors trackedECAD

Recurring kicad-studio-kit (oaslananka) errors, by impact

  1. 1a KiCad manufacturing release gate checked the CLI version in a preflight (1.37.0) and then accepted a green readiness result from 1.68.3 and approved the manufacturing evidence anyway; a regression test proved it, and now a version mismatch fails closed.ECAD · Agent failure · 1 report66
  2. 2a KiCad manufacturing release gate's review preview silently accepted fields the schema forbids, including private source evidence.ECAD · Agent failure · 1 report66
  3. 3A KiCad VS Code extension blocked a multi-sheet BOM export correctly, then told the user to check the wrong thing. When the configured KiCad CLI lacked native multi-sheet BOM, the message said to make sure the schematic opens in KiCad. The fix names the CLI incompatibility and still refuses to fall back to a partial offline parse of a hierarchical BOM. It is the second BOM-export message fix in this repo in two hours.ECAD · Agent failure · 1 report66
  4. 4A KiCad VS Code extension could show a partial BOM, or call a cancelled export a success, after the KiCad CLI went missing. A CLI exiting after writing half a temporary CSV, a launcher disappearing, or a user hitting Cancel all got the same generic 'verify the project opens in KiCad' message.ECAD · Agent failure · 1 report66
  5. 5A VS Code KiCad extension's DRC copy brought the library tables but not the local libraries, so real projects got fake missing-footprint warnings. Repeated courtyard overlaps also collapsed into fewer rows when their message matched. After the fix, DRC matched KiCad's CLI on three real boards (90/90, 138/138, 7/7).ECAD · Agent failure · 1 report66
  6. 6Auto-selected project precedence meant that in a multi-project workspace, ERC findings could show for a schematic other than the open one. The fix keeps only findings with verified KiCad UUID locations and drops stale ones.ECAD · Agent failure · 1 report66
  7. 7In a KiCad VS Code extension, 'Export PDF (PCB)' silently fails on KiCad 10.0.8 because it passes --layers ALL and the CLI now rejects 'ALL'. The schematic and board PDFs also wrote to the same <project>.pdf, so one could overwrite the other. Fixed by discovering the board's real layers, inner copper included, and naming the board PDF separately. This is the second KiCad 10.0.8 format break in this repo today after the 7:41 nets one.ECAD · Agent failure · 1 report66
  8. 8KiCad Studio for VS Code exported an incomplete BOM for multi-sheet schematics: a lightweight parser was used on hierarchical designs, and exports could come from a stale text tab instead of the editor in front. The fix prefers the real KiCad CLI for multi-sheet designs and refuses an incomplete BOM. On a 59-part demo the references now match the CLI exactly.ECAD · Agent failure · 1 report66
  9. 9KiCad Studio told users their schematic was broken when the real problem was a read-only output folder for the BOM CSV. The fix matches EACCES/EPERM against the configured output path only, so an unreadable input schematic is not blamed on the output. Tested with KiCad CLI 10.0.8 on a real multi-sheet project, and BOM references still match 59/59.ECAD · Agent failure · 1 report66
  10. 10Real KiCad 10.0.8 boards saved with the new named-net syntax showed 0 nets in the VS Code viewer while pcbnew reported 71. The fix rewrites only the in-memory renderer copy and never touches the design on disk. A cancelled kicad-cli DRC run also came back as a generic AbortError instead of a user cancel.ECAD · Agent failure · 1 report66
  11. 11Run Design Rule Check was missing from the VS Code Command Palette even with a real KiCad 10 PCB open, because the palette filter depended on an editor context key that was not reliably set. The unmodified 2.5.0 candidate still failed palette discovery, so the release was held.ECAD · Agent failure · 1 report66
  12. 12The net inspector could say a board had no nets when the renderer had not finished loading, and reference search silently succeeded on missing parts.ECAD · Agent failure · 1 report66
  13. 13The PCB diff report showed 0 to 0 tracks, vias and zones on real KiCad boards because KiCad writes (segment, (via and (zone with a newline before the fields and the counter expected a space. Per-layer counts were populated in the same report. On the test board the real change was 1,113 to 1,112 tracks.ECAD · Agent failure · 1 report66
  14. 14The viewer used to report Focused reference for any string; now it checks placed symbols, rejects nonexistent ones, and says a reference exists in source but cannot be focused in the static SVG fallback.ECAD · Agent failure · 1 report66

How it fails

  1. Agent failure14

Engineering tools that address these errors

Sourced from public community reports. Costs and delays on individual entries are modeled estimates, not measurements.