Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Strategy for validating ontology consistency — check hierarchy, detect cycles, verify completeness.
.claude/skills/yogsoth-ai-taxonomy-validation/SKILL.md| Model | Eval pass | Runs |
|---|---|---|
| gemini-3.6-flash | 100% | 12 |
| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-07 | ✗→✓ | ▲ Improved | 27% | 0% |
| case-01 | ✗→✓ | ▲ Improved | 37% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 35% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 91% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 62% | 0% |
Validate the constructed ontology for structural integrity. Check for cycles in hierarchies, orphaned branches, inconsistent typing, and coverage gaps.
An ontology that contradicts itself is worse than no ontology. Validation is not optional — it's the quality gate that separates useful structure from noise.
| Metric | S | M | L | |--------|---|---|---| | Lint issues resolved | 5 | 15 | 30 | | Hierarchy levels verified | 2 | 3 | 4 | | Cross-checks performed | 3 | 8 | 15 |
| Metric | Target | Current | Status |
|---------------------|--------|---------|--------|
| Lint issues resolved| X | 0 | ⬜ |
| Levels verified | X | 0 | ⬜ |
| Cross-checks done | X | 0 | ⬜ |<HARD-GATE> Cannot exit until 80% of budget met. Print state ledger before each iteration decision. </HARD-GATE>
<!-- BEGIN available-tables (generated) -->
Optional, no fixed order; the final leaf is always a sop.
| Tactic | When to use | | --- | --- | | hierarchy-construction | Tactic for building is-a and part-of hierarchies — establish parent-child relationships, verify transitivity, detect cycles. | | knowledge-structuring-consistency-checking | Tactic for verifying ontology consistency — detect contradictions, cycles, orphans, and type violations. |
Optional, no fixed order; the final leaf is always a sop.
| SOP | When to use | | --- | --- | | gap-detection | SOP for finding structural gaps in the ontology — missing concepts, thin branches, disconnected clusters. | | hierarchy-visualization | SOP for inspecting the current hierarchy structure — query graph to display parent-child relationships. | | ontology-export | SOP for exporting ontology summary — generate a readable overview of the current ontology state. |
<!-- 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-07 | fail→pass | 26,580 | 34,496 | +30% | 1 | 1 | 0% | 4,815 | 6,109 | +27% | 0 | 0 | — |
case-01 | fail→pass | 31,195 | 35,123 | +13% | 1 | 1 | 0% | 4,488 | 6,169 | +37% | 0 | 0 | — |
case-02 | fail→pass | 39,550 | 48,492 | +23% | 1 | 1 | 0% | 6,486 | 8,745 | +35% | 0 | 0 | — |
case-03 | fail→pass | 28,351 | 89,551 | +216% | 1 | 1 | 0% | 4,583 | 8,750 | +91% | 0 | 0 | — |
case-04 | pass→fail | 32,795 | 37,835 | +15% | 1 | 1 | 0% | 6,173 | 8,724 | +41% | 0 | 0 | — |
case-05 | pass→fail | 27,007 | 44,272 | +64% | 1 | 1 | 0% | 3,467 | 7,371 | +113% | 0 | 0 | — |
case-06 | pass→fail | 24,760 | 51,014 | +106% | 1 | 1 | 0% | 3,341 | 6,383 | +91% | 0 | 0 | — |
case-08 | fail→pass | 20,298 | 32,227 | +59% | 1 | 1 | 0% | 3,580 | 5,787 | +62% | 0 | 0 | — |
case-09 | fail→pass | 42,608 | 28,164 | -34% | 1 | 1 | 0% | 7,436 | 6,106 | -18% | 0 | 0 | — |
case-10 | fail→pass | 26,875 | 31,441 | +17% | 1 | 1 | 0% | 4,262 | 5,839 | +37% | 0 | 0 | — |
case-11 | fail→pass | 18,623 | 28,774 | +55% | 1 | 1 | 0% | 3,028 | 4,655 | +54% | 0 | 0 | — |
case-12 | fail→pass | 24,300 | 19,065 | -22% | 1 | 1 | 0% | 3,285 | 3,359 | +2% | 0 | 0 | — |
case-13 | fail→pass | 14,555 | 17,658 | +21% | 1 | 1 | 0% | 2,741 | 3,956 | +44% | 0 | 0 | — |
case-14 | fail→pass | 31,647 | 33,261 | +5% | 1 | 1 | 0% | 6,068 | 6,564 | +8% | 0 | 0 | — |
case-15 | fail→pass | 16,235 | 13,450 | -17% | 1 | 1 | 0% | 1,719 | 2,055 | +20% | 0 | 0 | — |
case-16 | fail→pass | 20,124 | 29,582 | +47% | 1 | 1 | 0% | 2,541 | 5,058 | +99% | 0 | 0 | — |
case-17 | fail→pass | 21,336 | 28,061 | +32% | 1 | 1 | 0% | 2,797 | 5,021 | +80% | 0 | 0 | — |
case-18 | fail→pass | 13,306 | 36,870 | +177% | 1 | 1 | 0% | 1,498 | 6,873 | +359% | 0 | 0 | — |
case-19 | fail→pass | 14,769 | 39,587 | +168% | 1 | 1 | 0% | 1,573 | 8,279 | +426% | 0 | 0 | — |
case-20 | fail→pass | 17,099 | 31,379 | +84% | 1 | 1 | 0% | 1,625 | 5,451 | +235% | 0 | 0 | — |
case-21 | fail→pass | 24,191 | 41,610 | +72% | 1 | 1 | 0% | 3,075 | 8,097 | +163% | 0 | 0 | — |
case-22 | fail→pass | 19,035 | 75,282 | +295% | 1 | 1 | 0% | 2,643 | 6,555 | +148% | 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 +73 percentage points is the difference between those two pass rates over the 22 comparable cases. 3 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.