Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Auri: assistente de voz inteligente (Alexa + Claude claude-opus-4-20250805). Visao do produto, persona Vitoria Neural, stack AWS, modelo Free/Pro/Business/Enterprise, roadmap 4 fases, GTM, north star WAC e analise competitiva.
.claude/skills/sickn33-auri-core/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-15 | ✗→✓ | ▲ Improved | 208% | 0% |
| case-01 | ✗→✓ | ▲ Improved | 151% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 121% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 280% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 161% | 0% |
Read the detailed guide before executing this skill. It retains the complete procedure and reference material. Treat its safety, prerequisites, and validation requirements as mandatory. For focused work, load the relevant sections; for end-to-end work, read the guide completely.
User request:
> Review this product request with the Auri workflow and return the documented product deliverable.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-15 | fail→pass | 11,272 | 4,316 | -62% | 1 | 1 | 0% | 2,307 | 7,109 | +208% | 0 | 0 | — |
case-01 | fail→pass | 14,615 | 11,055 | -24% | 1 | 1 | 0% | 3,405 | 8,535 | +151% | 0 | 0 | — |
case-02 | fail→pass | 21,650 | 9,089 | -58% | 1 | 1 | 0% | 3,674 | 8,114 | +121% | 0 | 0 | — |
case-03 | fail→fail | 24,375 | 22,865 | -6% | 1 | 1 | 0% | 4,169 | 9,801 | +135% | 0 | 0 | — |
case-04 | pass→pass | 9,368 | 8,999 | -4% | 1 | 1 | 0% | 2,008 | 8,140 | +305% | 0 | 0 | — |
case-05 | pass→pass | 4,840 | 3,635 | -25% | 1 | 1 | 0% | 1,063 | 7,072 | +565% | 0 | 0 | — |
case-06 | pass→pass | 13,841 | 11,382 | -18% | 1 | 1 | 0% | 2,505 | 8,576 | +242% | 0 | 0 | — |
case-07 | fail→pass | 12,755 | 10,183 | -20% | 1 | 1 | 0% | 2,168 | 8,229 | +280% | 0 | 0 | — |
case-08 | fail→pass | 19,644 | 13,098 | -33% | 1 | 1 | 0% | 3,404 | 8,889 | +161% | 0 | 0 | — |
case-09 | fail→fail | 15,137 | 11,716 | -23% | 1 | 1 | 0% | 3,995 | 9,023 | +126% | 0 | 0 | — |
case-10 | fail→pass | 7,781 | 2,955 | -62% | 1 | 1 | 0% | 1,513 | 6,977 | +361% | 0 | 0 | — |
case-11 | fail→pass | 15,007 | 10,633 | -29% | 1 | 1 | 0% | 2,552 | 8,390 | +229% | 0 | 0 | — |
case-12 | fail→pass | 7,681 | 3,880 | -49% | 1 | 1 | 0% | 1,251 | 7,071 | +465% | 0 | 0 | — |
case-13 | fail→pass | 9,001 | 5,439 | -40% | 1 | 1 | 0% | 1,572 | 7,366 | +369% | 0 | 0 | — |
case-14 | fail→pass | 2,841 | 4,896 | +72% | 1 | 1 | 0% | 451 | 7,326 | +1524% | 0 | 0 | — |
case-16 | pass→pass | 12,716 | 3,602 | -72% | 1 | 1 | 0% | 2,188 | 7,045 | +222% | 0 | 0 | — |
case-17 | fail→pass | 14,735 | 11,571 | -21% | 1 | 1 | 0% | 2,325 | 8,275 | +256% | 0 | 0 | — |
case-18 | fail→pass | 11,597 | 10,942 | -6% | 1 | 1 | 0% | 2,204 | 8,733 | +296% | 0 | 0 | — |
case-19 | fail→fail | 18,928 | 11,323 | -40% | 1 | 1 | 0% | 3,304 | 8,438 | +155% | 0 | 0 | — |
case-20 | fail→pass | 14,908 | 10,098 | -32% | 1 | 1 | 0% | 2,483 | 8,281 | +234% | 0 | 0 | — |
case-21 | fail→pass | 9,382 | 2,686 | -71% | 1 | 1 | 0% | 1,947 | 6,895 | +254% | 0 | 0 | — |
case-22 | fail→pass | 10,336 | 3,230 | -69% | 1 | 1 | 0% | 2,141 | 6,990 | +226% | 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 +68 percentage points is the difference between those two pass rates over the 22 comparable cases.
The publisher has shipped newer versions since this run, so these numbers describe v1, not the version currently listed.
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.