Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Systematic coverage evaluation pipeline — benchmark inventory, method-problem crossing, and intersection evaluation to map explored vs unexplored solution space.
.claude/skills/yogsoth-ai-coverage-analysis/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-06 | ✗→✓ | ▲ Improved | 9% | 0% |
| case-01 | ✗→✓ | ▲ Improved | 3% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 129% | 0% |
| case-08 | ✗→✓ | ▲ Improved | -19% | 0% |
| case-10 | ✗→✓ | ▲ Improved | -29% | 0% |
Systematic coverage evaluation: inventory all known methods, cross them against problems, and evaluate which intersections remain unexplored.
Run benchmark-inventory SOP to catalog all known solutions/methods in the target domain. Output: structured method inventory with performance, applicability, and limitations.
Run method-problem-crossing SOP to build a cross-reference matrix of methods × problems. Output: complete matrix with all cells populated.
Run intersection-evaluation SOP to annotate each matrix cell as explored, partial, or unexplored. Output: annotated matrix with exploration status and priority ranking.
| Metric | Floor | |--------|-------| | Methods cataloged | ≥80% of known methods in domain | | Problems enumerated | ≥80% of known problems in domain | | Matrix cells evaluated | 100% of M×P cells | | Unexplored intersections identified | ≥5 | | Priority ranking produced | yes |
| SOP | Role | |-----|------| | benchmark-inventory | Stage 1 — catalog all known solutions | | method-problem-crossing | Stage 2 — build cross-reference matrix | | intersection-evaluation | Stage 3 — annotate exploration status |
<!-- BEGIN available-tables (generated) -->
Optional, no fixed order; the final leaf is always a sop.
| SOP | When to use | | --- | --- | | creative-ideation-benchmark-inventory | Catalog all known solutions/methods in a domain with performance, applicability, and limitations. | | intersection-evaluation | Evaluate exploration status of each cell in a method×problem matrix, annotating as explored, partial, or unexplored. | | method-problem-crossing | Build method×problem cross-reference matrix showing which methods have been applied to which problems. |
<!-- 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-04 | pass→pass | 26,073 | 30,121 | +16% | 1 | 1 | 0% | 4,153 | 5,417 | +30% | 0 | 0 | — |
case-05 | pass→pass | 28,668 | 30,026 | +5% | 1 | 1 | 0% | 5,667 | 6,602 | +16% | 0 | 0 | — |
case-06 | fail→pass | 20,752 | 18,255 | -12% | 1 | 1 | 0% | 3,309 | 3,593 | +9% | 0 | 0 | — |
case-01 | fail→pass | 35,881 | 34,473 | -4% | 1 | 1 | 0% | 6,211 | 6,415 | +3% | 0 | 0 | — |
case-02 | fail→pass | 14,258 | 29,404 | +106% | 1 | 1 | 0% | 2,899 | 6,639 | +129% | 0 | 0 | — |
case-03 | pass→pass | 31,091 | 30,717 | -1% | 1 | 1 | 0% | 5,860 | 6,167 | +5% | 0 | 0 | — |
case-07 | pass→fail | 67,927 | 33,335 | -51% | 1 | 1 | 0% | 6,196 | 6,213 | +0% | 0 | 0 | — |
case-08 | fail→pass | 36,275 | 26,751 | -26% | 1 | 1 | 0% | 6,176 | 4,996 | -19% | 0 | 0 | — |
case-09 | fail→fail | 39,675 | 35,320 | -11% | 1 | 1 | 0% | 6,191 | 6,210 | +0% | 0 | 0 | — |
case-10 | fail→pass | 38,083 | 24,160 | -37% | 1 | 1 | 0% | 5,919 | 4,186 | -29% | 0 | 0 | — |
case-11 | pass→pass | 22,117 | 20,489 | -7% | 1 | 1 | 0% | 3,868 | 3,976 | +3% | 0 | 0 | — |
case-12 | fail→fail | 27,918 | 20,669 | -26% | 1 | 1 | 0% | 4,290 | 3,670 | -14% | 0 | 0 | — |
case-13 | fail→pass | 34,613 | 34,233 | -1% | 1 | 1 | 0% | 6,170 | 6,602 | +7% | 0 | 0 | — |
case-14 | fail→fail | 17,073 | 43,186 | +153% | 1 | 1 | 0% | 2,769 | 5,024 | +81% | 0 | 0 | — |
case-15 | fail→pass | 36,029 | 21,130 | -41% | 1 | 1 | 0% | 6,172 | 3,928 | -36% | 0 | 0 | — |
case-16 | fail→pass | 38,127 | 99,863 | +162% | 1 | 1 | 0% | 6,177 | 6,609 | +7% | 0 | 0 | — |
case-17 | fail→fail | 27,041 | 33,033 | +22% | 1 | 1 | 0% | 4,287 | 5,973 | +39% | 0 | 0 | — |
case-18 | fail→pass | 36,684 | 38,154 | +4% | 1 | 1 | 0% | 6,171 | 6,603 | +7% | 0 | 0 | — |
case-19 | fail→pass | 39,193 | 37,943 | -3% | 1 | 1 | 0% | 6,175 | 6,607 | +7% | 0 | 0 | — |
case-20 | fail→pass | 36,164 | 25,306 | -30% | 1 | 1 | 0% | 5,665 | 4,498 | -21% | 0 | 0 | — |
case-21 | pass→pass | 35,378 | 27,384 | -23% | 1 | 1 | 0% | 5,635 | 4,731 | -16% | 0 | 0 | — |
case-22 | fail→pass | 26,445 | 35,828 | +35% | 1 | 1 | 0% | 3,998 | 5,974 | +49% | 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 +50 percentage points is the difference between those two pass rates over the 22 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.