Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Validate analogy depth and transfer viability. Ensures only deep structural analogies (not surface-level similarities) proceed to transfer.
.claude/skills/yogsoth-ai-bridge-validation/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | -9% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 21% | 0% |
| case-10 | ✗→✓ | ▲ Improved | -28% | 0% |
| case-11 | ✗→✓ | ▲ Improved | 64% | 0% |
| case-17 | ✗→✓ | ▲ Improved | -34% | 0% |
Validate analogy depth and transfer viability before committing resources to transfer.
Apply analogy-quality-assessment SOP to each candidate analogy. Classify depth:
For accepted analogies, test transfer viability using transfer-adaptation SOP:
Verify the adapted transfer maintains structural consistency:
| Metric | Floor | |--------|-------| | Analogies assessed | all candidates | | Validated deep analogies | ≥2 | | Transfer viability confirmed | ≥2 | | Structural consistency verified | ≥2 |
| SOP | Role | |-----|------| | analogy-quality-assessment | Stage 1 — classify analogy depth | | transfer-adaptation | Stage 2 — test and adapt transfer | | structural-mapping | Stage 3 — verify structural consistency | | abstraction-extraction | Support — re-abstract if mapping fails |
<!-- BEGIN available-tables (generated) -->
Optional, no fixed order; the final leaf is always a sop.
| SOP | When to use | | --- | --- | | analogy-quality-assessment | Assess analogy depth (surface/structural/systemic). Determines whether an analogy warrants transfer investment. | | structural-mapping | Map source→target structural correspondences. Identifies corresponding, missing, and extra elements between domains. | | transfer-adaptation | Adapt transferred principle to target problem constraints. Produces concrete adapted solutions from abstract principles. |
<!-- 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-01 | fail→pass | 39,531 | 33,393 | -16% | 1 | 1 | 0% | 5,953 | 5,428 | -9% | 0 | 0 | — |
case-02 | fail→pass | 32,422 | 34,960 | +8% | 1 | 1 | 0% | 4,684 | 5,688 | +21% | 0 | 0 | — |
case-03 | pass→pass | 24,620 | 27,099 | +10% | 1 | 1 | 0% | 4,009 | 4,793 | +20% | 0 | 0 | — |
case-04 | pass→pass | 12,906 | 15,186 | +18% | 1 | 1 | 0% | 2,221 | 3,104 | +40% | 0 | 0 | — |
case-05 | pass→fail | 17,248 | 36,866 | +114% | 1 | 1 | 0% | 2,768 | 6,241 | +125% | 0 | 0 | — |
case-06 | pass→pass | 19,074 | 32,319 | +69% | 1 | 1 | 0% | 3,894 | 6,531 | +68% | 0 | 0 | — |
case-07 | pass→pass | 11,957 | 16,096 | +35% | 1 | 1 | 0% | 1,705 | 2,866 | +68% | 0 | 0 | — |
case-08 | fail→fail | 16,275 | 15,012 | -8% | 1 | 1 | 0% | 2,201 | 2,562 | +16% | 0 | 0 | — |
case-09 | fail→fail | 19,508 | 19,937 | +2% | 1 | 1 | 0% | 2,603 | 3,332 | +28% | 0 | 0 | — |
case-10 | fail→pass | 24,009 | 14,524 | -40% | 1 | 1 | 0% | 3,530 | 2,528 | -28% | 0 | 0 | — |
case-11 | fail→pass | 20,575 | 30,801 | +50% | 1 | 1 | 0% | 3,023 | 4,959 | +64% | 0 | 0 | — |
case-12 | pass→pass | 24,927 | 24,071 | -3% | 1 | 1 | 0% | 3,834 | 4,403 | +15% | 0 | 0 | — |
case-13 | pass→pass | 31,239 | 26,178 | -16% | 1 | 1 | 0% | 4,549 | 4,284 | -6% | 0 | 0 | — |
case-14 | pass→pass | 15,087 | 18,891 | +25% | 1 | 1 | 0% | 2,194 | 3,221 | +47% | 0 | 0 | — |
case-15 | pass→pass | 15,421 | 23,903 | +55% | 1 | 1 | 0% | 2,224 | 4,091 | +84% | 0 | 0 | — |
case-16 | fail→fail | 22,462 | 23,351 | +4% | 1 | 1 | 0% | 3,064 | 3,717 | +21% | 0 | 0 | — |
case-17 | fail→pass | 10,862 | 3,122 | -71% | 1 | 1 | 0% | 1,552 | 1,023 | -34% | 0 | 0 | — |
case-18 | fail→pass | 11,607 | 3,421 | -71% | 1 | 1 | 0% | 1,807 | 1,042 | -42% | 0 | 0 | — |
case-19 | pass→pass | 17,327 | 10,861 | -37% | 1 | 1 | 0% | 2,458 | 2,097 | -15% | 0 | 0 | — |
case-20 | pass→pass | 14,370 | 13,139 | -9% | 1 | 1 | 0% | 2,043 | 2,427 | +19% | 0 | 0 | — |
case-21 | pass→pass | 22,549 | 24,273 | +8% | 1 | 1 | 0% | 3,413 | 3,975 | +16% | 0 | 0 | — |
case-22 | pass→pass | 19,785 | 12,047 | -39% | 1 | 1 | 0% | 2,776 | 2,277 | -18% | 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. 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.