Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Tactic: Select the most suitable RQ framework and apply it systematically
.claude/skills/yogsoth-ai-framework-selection-and-application/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-02 | ✗→✓ | ▲ Improved | -17% | 0% |
| case-01 | ✗→✓ | ▲ Improved | -35% | 0% |
| case-04 | ✗→✓ | ▲ Improved | -31% | 0% |
| case-09 | ✗→✓ | ▲ Improved | -3% | 0% |
| case-10 | ✗→✓ | ▲ Improved | -28% | 0% |
Select the most suitable RQ framework and apply it systematically — from framework matching to full population.
Decompose "structuring a research question with a framework" into: first select the framework → then apply the framework. Selection and application are two independent steps, handled by different SOPs.
| SOP | Responsibility | When to call | |-----|------|---------| | framework-matching | Match the most suitable framework based on research type | First step, always called | | pico-application | Apply the PICO/PICOTS framework | Quantitative/intervention research | | spider-application | Apply the SPIDER framework | Qualitative research | | spice-application | Apply the SPICE framework | Evaluation research | | eclipse-application | Apply the ECLIPSE framework | Mixed methods research |
Standard pattern:
Comparison pattern (M/L tier):
After execution completes, report:
<!-- BEGIN available-tables (generated) -->
Optional, no fixed order; the final leaf is always a sop.
| SOP | When to use | | --- | --- | | eclipse-application | SOP: apply the ECLIPSE framework to structure a mixed-methods research question | | framework-matching | SOP: Match the most suitable RQ framework based on research type | | pico-application | SOP: apply the PICO/PICOTS framework to structure a research question | | spice-application | SOP: Apply the SPICE framework to structure an evaluation research question | | spider-application | SOP: Apply the SPIDER framework to structure a qualitative research question |
<!-- END available-tables (generated) -->
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-02 | fail→pass | 12,957 | 9,150 | -29% | 1 | 1 | 0% | 2,381 | 1,969 | -17% | 0 | 0 | — |
case-01 | fail→pass | 25,661 | 12,767 | -50% | 1 | 1 | 0% | 4,221 | 2,756 | -35% | 0 | 0 | — |
case-03 | pass→pass | 12,942 | 12,224 | -6% | 1 | 1 | 0% | 2,237 | 2,608 | +17% | 0 | 0 | — |
case-04 | fail→pass | 18,094 | 10,596 | -41% | 1 | 1 | 0% | 3,328 | 2,308 | -31% | 0 | 0 | — |
case-05 | pass→pass | 17,477 | 7,807 | -55% | 1 | 1 | 0% | 3,228 | 1,895 | -41% | 0 | 0 | — |
case-06 | pass→pass | 16,527 | 6,894 | -58% | 1 | 1 | 0% | 2,956 | 1,678 | -43% | 0 | 0 | — |
case-07 | pass→pass | 12,624 | 13,265 | +5% | 1 | 1 | 0% | 2,166 | 2,329 | +8% | 0 | 0 | — |
case-08 | pass→pass | 22,873 | 10,790 | -53% | 1 | 1 | 0% | 3,713 | 2,370 | -36% | 0 | 0 | — |
case-09 | fail→pass | 14,538 | 10,467 | -28% | 1 | 1 | 0% | 2,387 | 2,318 | -3% | 0 | 0 | — |
case-10 | fail→pass | 16,828 | 8,216 | -51% | 1 | 1 | 0% | 2,741 | 1,986 | -28% | 0 | 0 | — |
case-11 | fail→pass | 20,212 | 19,210 | -5% | 1 | 1 | 0% | 3,305 | 3,938 | +19% | 0 | 0 | — |
case-12 | fail→pass | 10,450 | 10,187 | -3% | 1 | 1 | 0% | 1,770 | 2,231 | +26% | 0 | 0 | — |
case-13 | fail→pass | 11,693 | 15,805 | +35% | 1 | 1 | 0% | 1,904 | 2,745 | +44% | 0 | 0 | — |
case-14 | fail→pass | 26,861 | 16,511 | -39% | 1 | 1 | 0% | 4,439 | 3,532 | -20% | 0 | 0 | — |
case-15 | pass→pass | 18,366 | 6,417 | -65% | 1 | 1 | 0% | 3,204 | 1,695 | -47% | 0 | 0 | — |
case-16 | fail→pass | 11,799 | 8,340 | -29% | 1 | 1 | 0% | 2,088 | 1,882 | -10% | 0 | 0 | — |
case-17 | fail→pass | 30,225 | 15,502 | -49% | 1 | 1 | 0% | 4,713 | 3,286 | -30% | 0 | 0 | — |
case-18 | fail→pass | 38,448 | 12,830 | -67% | 1 | 1 | 0% | 6,174 | 2,662 | -57% | 0 | 0 | — |
case-19 | pass→pass | 12,303 | 9,350 | -24% | 1 | 1 | 0% | 2,305 | 2,207 | -4% | 0 | 0 | — |
case-20 | pass→fail | 13,452 | 17,155 | +28% | 1 | 1 | 0% | 2,726 | 3,740 | +37% | 0 | 0 | — |
case-21 | fail→fail | 11,712 | 20,245 | +73% | 1 | 1 | 0% | 2,058 | 4,097 | +99% | 0 | 0 | — |
case-22 | pass→fail | 18,235 | 16,815 | -8% | 1 | 1 | 0% | 3,245 | 3,665 | +13% | 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 +45 percentage points is the difference between those two pass rates over the 22 comparable cases. 2 cases got worse with the skill loaded, and they are 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.