Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Design and implement Arduino integration with Azure IoT Hub and IoT Edge, including secure provisioning, resilient telemetry, command handling, and production guardrails.
.claude/skills/arduino-azure-iot-edge-integration/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-03 | ✗→✓ | ▲ Improved | 1% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 25% | 0% |
| case-09 | ✗→✓ | ▲ Improved | 34% | 0% |
| case-15 | ✗→✓ | ▲ Improved | 18% | 0% |
| case-22 | ✗→✓ | ▲ Improved | 27% | 0% |
Use this skill when the user needs to connect Arduino-class devices to Azure IoT, especially in edge-heavy scenarios (gateways, intermittent networks, offline buffering, and local actuation).
Use this skill for requests such as:
Before recommending an IoT Edge topology or runtime behavior, review:
If documentation cannot be consulted, proceed with explicit assumptions and highlight them in a dedicated section.
Before proposing firmware, wiring, or communication implementation details, consult official Arduino sources first:
When choosing between implementation alternatives, prioritize official Arduino guidance over community snippets unless there is a clear technical reason to deviate.
Use when connectivity is stable and cloud latency is acceptable.
Use when links are constrained, local control is required, or batching improves cost/reliability.
Define:
Require:
Plan and document:
Define routes for:
Specify minimum operations telemetry:
When relevant, combine with:
azure-smart-city-iot-solution-builder for city-wide architecture and phased rollout.azure-resource-visualizer for relationship diagrams.appinsights-instrumentation for app and service telemetry patterns.Also use references/arduino-official-best-practices.md as a quality baseline for firmware and hardware recommendations, and references/arduino-iot-checklist.md before finalizing architecture or implementation guidance.
Always provide:
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-05 | pass→pass | 11,111 | 14,742 | +33% | 1 | 1 | 0% | 2,423 | 4,212 | +74% | 0 | 0 | — |
case-06 | pass→pass | 21,393 | 24,246 | +13% | 1 | 1 | 0% | 5,239 | 6,696 | +28% | 0 | 0 | — |
case-03 | fail→pass | 29,511 | 24,562 | -17% | 1 | 1 | 0% | 5,738 | 5,814 | +1% | 0 | 0 | — |
case-04 | pass→pass | 18,778 | 20,196 | +8% | 1 | 1 | 0% | 3,759 | 5,024 | +34% | 0 | 0 | — |
case-12 | pass→pass | 14,855 | 14,405 | -3% | 1 | 1 | 0% | 3,018 | 3,989 | +32% | 0 | 0 | — |
case-01 | pass→pass | 20,270 | 20,661 | +2% | 1 | 1 | 0% | 3,554 | 5,064 | +42% | 0 | 0 | — |
case-02 | pass→pass | 20,370 | 21,457 | +5% | 1 | 1 | 0% | 3,552 | 5,011 | +41% | 0 | 0 | — |
case-07 | pass→pass | 11,948 | 14,111 | +18% | 1 | 1 | 0% | 2,354 | 3,718 | +58% | 0 | 0 | — |
case-08 | fail→pass | 15,536 | 14,512 | -7% | 1 | 1 | 0% | 2,968 | 3,721 | +25% | 0 | 0 | — |
case-09 | fail→pass | 19,336 | 25,239 | +31% | 1 | 1 | 0% | 4,494 | 6,005 | +34% | 0 | 0 | — |
case-10 | pass→pass | 17,195 | 16,653 | -3% | 1 | 1 | 0% | 3,137 | 4,399 | +40% | 0 | 0 | — |
case-11 | pass→pass | 15,097 | 19,809 | +31% | 1 | 1 | 0% | 2,743 | 4,439 | +62% | 0 | 0 | — |
case-13 | fail→fail | 15,293 | 13,455 | -12% | 1 | 1 | 0% | 2,776 | 3,480 | +25% | 0 | 0 | — |
case-14 | pass→pass | 12,918 | 20,304 | +57% | 1 | 1 | 0% | 2,330 | 4,670 | +100% | 0 | 0 | — |
case-15 | fail→pass | 15,186 | 12,910 | -15% | 1 | 1 | 0% | 2,806 | 3,318 | +18% | 0 | 0 | — |
case-16 | pass→pass | 14,352 | 17,133 | +19% | 1 | 1 | 0% | 2,897 | 4,138 | +43% | 0 | 0 | — |
case-17 | pass→pass | 12,252 | 15,701 | +28% | 1 | 1 | 0% | 2,435 | 3,989 | +64% | 0 | 0 | — |
case-18 | pass→pass | 12,435 | 6,960 | -44% | 1 | 1 | 0% | 2,440 | 2,401 | -2% | 0 | 0 | — |
case-19 | pass→pass | 12,726 | 14,097 | +11% | 1 | 1 | 0% | 2,402 | 3,693 | +54% | 0 | 0 | — |
case-20 | pass→pass | 15,537 | 15,652 | +1% | 1 | 1 | 0% | 3,060 | 4,027 | +32% | 0 | 0 | — |
case-21 | fail→fail | 15,328 | 15,181 | -1% | 1 | 1 | 0% | 3,086 | 3,908 | +27% | 0 | 0 | — |
case-22 | fail→pass | 12,437 | 10,839 | -13% | 1 | 1 | 0% | 2,442 | 3,112 | +27% | 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 +23 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.
| Model | Method | Date | Lift |
|---|---|---|---|
| gemini-3.6-flash | verified | 7/23/2026 | +27% |
Other measured skills in the registry, with their headline benchmark lift.