Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Debug WebSocket connection issues between CLI and FigJam plugin. Use when diagrams aren't syncing, connection fails, plugin shows "Connecting..." indefinitely, or patches aren't being applied to canvas.
.claude/skills/aiskillstore-debug-connection/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | -11% | 0% |
| case-02 | ✗→✓ | ▲ Improved | -11% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 16% | 0% |
| case-16 | ✗→✓ | ▲ Improved | -63% | 0% |
| case-08 | ✗→✓ | ▲ Improved | -39% | 0% |
CLI serve (Bun) ←── WebSocket ──→ Plugin UI (ui.ts) ←── postMessage ──→ Plugin Main (code.ts)
│ │ │
File watcher Browser APIs Figma API
YAML parsing WebSocket client Canvas rendering| Issue | Symptom | Solution | |-------|---------|----------| | CLI not running | "Connecting..." indefinitely | Start bun run packages/cli/src/index.ts serve diagram.yaml | | Port in use | Connection refused | Check lsof -i :3456, use different port | | Secret mismatch | Connects then immediately closes | Match --secret value in CLI and plugin | | YAML errors | Connected but no updates | Fix validation errors in CLI output | | docId mismatch | No response after hello | Ensure plugin docId matches YAML docId | | Patches not applied | Connected, canvas unchanged | Check Plugin Main console for render errors |
bun run packages/cli/src/index.ts build diagram.yamlPlugin CLI
│ │
│──── hello ─────────────────►│ {type:"hello", docId, secret?}
│◄──── full ─────────────────│ {type:"full", rev, ir}
│ │
│ [YAML changes] │
│◄──── patch ────────────────│ {type:"patch", baseRev, nextRev, ops}
│ │
│ [Reconnect] │
│──── requestFull ───────────►│ {type:"requestFull", docId}
│◄──── full ─────────────────│version, docId, nodesnodes as arraynodes as object (Record)bash# Start server bun run packages/cli/src/index.ts serve examples/diagram.yaml # Test with wscat wscat -c ws://localhost:3456 > {"type":"hello","docId":"test"}
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 20,564 | 15,052 | -27% | 1 | 1 | 0% | 3,593 | 3,205 | -11% | 0 | 0 | — |
case-02 | fail→pass | 17,598 | 11,208 | -36% | 1 | 1 | 0% | 2,707 | 2,404 | -11% | 0 | 0 | — |
case-03 | fail→pass | 18,559 | 16,864 | -9% | 1 | 1 | 0% | 2,993 | 3,457 | +16% | 0 | 0 | — |
case-04 | pass→pass | 16,378 | 12,932 | -21% | 1 | 1 | 0% | 2,906 | 2,787 | -4% | 0 | 0 | — |
case-05 | fail→fail | 15,162 | 14,222 | -6% | 1 | 1 | 0% | 3,022 | 3,217 | +6% | 0 | 0 | — |
case-06 | fail→fail | 11,520 | 18,908 | +64% | 1 | 1 | 0% | 2,041 | 2,642 | +29% | 0 | 0 | — |
case-16 | fail→pass | 15,383 | 2,135 | -86% | 1 | 1 | 0% | 2,348 | 865 | -63% | 0 | 0 | — |
case-07 | pass→pass | 6,879 | 3,566 | -48% | 1 | 1 | 0% | 958 | 1,139 | +19% | 0 | 0 | — |
case-08 | fail→pass | 9,519 | 2,198 | -77% | 1 | 1 | 0% | 1,507 | 915 | -39% | 0 | 0 | — |
case-09 | fail→pass | 9,790 | 1,866 | -81% | 1 | 1 | 0% | 1,428 | 840 | -41% | 0 | 0 | — |
case-10 | fail→pass | 48,245 | 2,108 | -96% | 1 | 1 | 0% | 6,172 | 883 | -86% | 0 | 0 | — |
case-11 | fail→pass | 9,247 | 2,114 | -77% | 1 | 1 | 0% | 1,276 | 931 | -27% | 0 | 0 | — |
case-12 | fail→pass | 14,919 | 2,286 | -85% | 1 | 1 | 0% | 2,266 | 896 | -60% | 0 | 0 | — |
case-13 | fail→pass | 12,468 | 2,863 | -77% | 1 | 1 | 0% | 2,029 | 1,006 | -50% | 0 | 0 | — |
case-14 | fail→fail | 10,832 | 2,293 | -79% | 1 | 1 | 0% | 1,690 | 975 | -42% | 0 | 0 | — |
case-15 | pass→pass | 10,771 | 2,035 | -81% | 1 | 1 | 0% | 1,957 | 863 | -56% | 0 | 0 | — |
case-17 | pass→pass | 10,795 | 4,118 | -62% | 1 | 1 | 0% | 1,435 | 1,191 | -17% | 0 | 0 | — |
case-18 | pass→pass | 6,088 | 2,876 | -53% | 1 | 1 | 0% | 916 | 935 | +2% | 0 | 0 | — |
case-19 | pass→pass | 4,742 | 2,304 | -51% | 1 | 1 | 0% | 734 | 911 | +24% | 0 | 0 | — |
case-20 | fail→pass | 6,235 | 4,472 | -28% | 1 | 1 | 0% | 1,184 | 1,337 | +13% | 0 | 0 | — |
case-21 | fail→pass | 8,948 | 3,166 | -65% | 1 | 1 | 0% | 1,469 | 1,090 | -26% | 0 | 0 | — |
case-22 | fail→pass | 10,981 | 1,873 | -83% | 1 | 1 | 0% | 1,782 | 813 | -54% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted. The headline lift of +59 percentage points is the difference between those two pass rates over the 22 comparable cases.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.