Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Expert patterns for Azure Functions development including isolated worker model, Durable Functions orchestration, cold start optimization, and production patterns. Covers .NET, Python, and Node.js programming models. Use when: azure function, azure functions, durable functions, azure serverless, function app.
.claude/skills/davila7-azure-functions/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | -24% | 0% |
| case-08 | ✓→✗ | ▼ Worse | -22% | 0% |
| case-15 | ✓→✓ | ＝ Same ✓ | -1% | 0% |
| case-16 | ✓→✓ | ＝ Same ✓ | -11% | 0% |
| case-02 | ✓→✓ | ＝ Same ✓ | -14% | 0% |
Modern .NET execution model with process isolation
Modern code-centric approach for TypeScript/JavaScript
Decorator-based approach for Python functions
| Issue | Severity | Solution | |-------|----------|----------| | Issue | high | ## Use async pattern with Durable Functions | | Issue | high | ## Use IHttpClientFactory (Recommended) | | Issue | high | ## Always use async/await | | Issue | medium | ## Configure maximum timeout (Consumption) | | Issue | high | ## Use isolated worker for new projects | | Issue | medium | ## Configure Application Insights properly | | Issue | medium | ## Check extension bundle (most common) | | Issue | medium | ## Add warmup trigger to initialize your code |
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 18,074 | 11,953 | -34% | 1 | 1 | 0% | 3,198 | 2,416 | -24% | 0 | 0 | — |
case-15 | pass→pass | 10,291 | 7,765 | -25% | 1 | 1 | 0% | 1,689 | 1,669 | -1% | 0 | 0 | — |
case-16 | pass→pass | 7,554 | 5,806 | -23% | 1 | 1 | 0% | 1,503 | 1,335 | -11% | 0 | 0 | — |
case-02 | pass→pass | 6,865 | 4,547 | -34% | 1 | 1 | 0% | 1,168 | 1,002 | -14% | 0 | 0 | — |
case-03 | pass→pass | 8,118 | 4,123 | -49% | 1 | 1 | 0% | 1,483 | 977 | -34% | 0 | 0 | — |
case-04 | pass→pass | 12,268 | 9,916 | -19% | 1 | 1 | 0% | 2,333 | 2,162 | -7% | 0 | 0 | — |
case-05 | pass→pass | 11,396 | 10,992 | -4% | 1 | 1 | 0% | 1,847 | 2,111 | +14% | 0 | 0 | — |
case-06 | pass→pass | 12,589 | 10,294 | -18% | 1 | 1 | 0% | 2,179 | 1,971 | -10% | 0 | 0 | — |
case-07 | pass→pass | 5,597 | 4,198 | -25% | 1 | 1 | 0% | 1,050 | 933 | -11% | 0 | 0 | — |
case-08 | pass→fail | 10,408 | 7,744 | -26% | 1 | 1 | 0% | 2,074 | 1,619 | -22% | 0 | 0 | — |
case-09 | pass→pass | 8,699 | 6,719 | -23% | 1 | 1 | 0% | 1,522 | 1,280 | -16% | 0 | 0 | — |
case-10 | fail→fail | 10,718 | 6,028 | -44% | 1 | 1 | 0% | 1,797 | 1,245 | -31% | 0 | 0 | — |
case-11 | pass→pass | 9,305 | 4,483 | -52% | 1 | 1 | 0% | 1,721 | 1,047 | -39% | 0 | 0 | — |
case-12 | pass→pass | 6,717 | 2,793 | -58% | 1 | 1 | 0% | 1,166 | 725 | -38% | 0 | 0 | — |
case-13 | pass→pass | 6,587 | 3,485 | -47% | 1 | 1 | 0% | 1,086 | 828 | -24% | 0 | 0 | — |
case-14 | pass→pass | 10,829 | 4,703 | -57% | 1 | 1 | 0% | 1,807 | 1,058 | -41% | 0 | 0 | — |
case-17 | pass→pass | 9,794 | 9,658 | -1% | 1 | 1 | 0% | 1,841 | 1,834 | -0% | 0 | 0 | — |
case-18 | pass→pass | 8,936 | 6,744 | -25% | 1 | 1 | 0% | 1,706 | 1,435 | -16% | 0 | 0 | — |
case-19 | pass→pass | 11,355 | 6,407 | -44% | 1 | 1 | 0% | 1,870 | 1,299 | -31% | 0 | 0 | — |
case-20 | pass→pass | 9,802 | 4,266 | -56% | 1 | 1 | 0% | 1,796 | 1,021 | -43% | 0 | 0 | — |
case-21 | pass→pass | 6,246 | 2,588 | -59% | 1 | 1 | 0% | 1,081 | 679 | -37% | 0 | 0 | — |
case-22 | pass→pass | 10,995 | 6,332 | -42% | 1 | 1 | 0% | 1,962 | 1,310 | -33% | 0 | 0 | — |
case-23 | pass→pass | 12,160 | 8,952 | -26% | 1 | 1 | 0% | 1,997 | 1,788 | -10% | 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. 23 cases were attempted. The headline lift of 0 percentage points is the difference between those two pass rates over the 23 comparable cases. 1 case got worse with the skill loaded, and it is included in that figure.
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.