Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Forward/backward patent citation and priority tracing until saturation
.claude/skills/yogsoth-ai-patent-family-tracing/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-02 | ✗→✓ | ▲ Improved | -2% | 0% |
| case-05 | ✗→✓ | ▲ Improved | -25% | 0% |
| case-06 | ✗→✓ | ▲ Improved | -27% | 0% |
| case-11 | ✗→✓ | ▲ Improved | -44% | 0% |
| case-04 | ✓→✓ | ＝ Same ✓ | -9% | 0% |
Traces patent citation networks and priority chains in both directions (forward citations, backward citations, priority links) until the search space saturates.
saturation-detection SOP to determine if expansion has converged| SOP | Role | |-----|------| | patent-query-formulation | Generate queries for citation lookup | | citation-network-analysis | Analyze the citation graph structure | | legal-status-assessment | Check status of discovered family members | | saturation-detection | Determine when expansion has converged |
Report to calling strategy after each execution:
<!-- BEGIN available-tables (generated) -->
Optional, no fixed order; the final leaf is always a sop.
| SOP | When to use | | --- | --- | | citation-network-analysis | Build and analyze patent citation networks — main path analysis, PageRank, cluster detection | | knowledge-acquisition-saturation-detection | Determine when additional searching yields diminishing returns. Analyzes the latest expansion batch against existing corpus to judge continue/near-saturation/saturated. Used by snowball and systematic-survey. | | legal-status-assessment | Determine patent legal status — active, expired, pending, lapsed, or revoked | | patent-query-formulation | Construct keyword + IPC/CPC + assignee combination search strategies for patent databases |
<!-- 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 | 20,435 | 13,465 | -34% | 1 | 1 | 0% | 1,966 | 1,782 | -9% | 0 | 0 | — |
case-01 | fail→fail | 44,794 | 57,813 | +29% | 1 | 1 | 0% | 6,636 | 7,907 | +19% | 0 | 0 | — |
case-02 | fail→pass | 57,200 | 43,710 | -24% | 1 | 1 | 0% | 7,307 | 7,134 | -2% | 0 | 0 | — |
case-03 | pass→pass | 32,399 | 19,980 | -38% | 1 | 1 | 0% | 2,366 | 1,568 | -34% | 0 | 0 | — |
case-05 | fail→pass | 40,192 | 30,936 | -23% | 1 | 1 | 0% | 2,449 | 1,843 | -25% | 0 | 0 | — |
case-06 | fail→pass | 16,286 | 34,209 | +110% | 1 | 1 | 0% | 1,776 | 1,290 | -27% | 0 | 0 | — |
case-07 | pass→pass | 12,450 | 19,756 | +59% | 1 | 1 | 0% | 1,260 | 1,641 | +30% | 0 | 0 | — |
case-08 | pass→pass | 12,750 | 10,776 | -15% | 1 | 1 | 0% | 1,150 | 1,568 | +36% | 0 | 0 | — |
case-09 | pass→pass | 15,459 | 8,487 | -45% | 1 | 1 | 0% | 1,670 | 1,077 | -36% | 0 | 0 | — |
case-10 | pass→pass | 21,441 | 8,946 | -58% | 1 | 1 | 0% | 1,826 | 967 | -47% | 0 | 0 | — |
case-11 | fail→pass | 17,987 | 7,727 | -57% | 1 | 1 | 0% | 1,700 | 960 | -44% | 0 | 0 | — |
case-12 | pass→pass | 16,033 | 8,455 | -47% | 1 | 1 | 0% | 1,696 | 1,115 | -34% | 0 | 0 | — |
case-13 | pass→pass | 16,755 | 9,007 | -46% | 1 | 1 | 0% | 1,792 | 1,210 | -32% | 0 | 0 | — |
case-14 | pass→pass | 21,168 | 11,908 | -44% | 1 | 1 | 0% | 2,736 | 1,690 | -38% | 0 | 0 | — |
case-15 | pass→pass | 15,018 | 8,814 | -41% | 1 | 1 | 0% | 1,446 | 1,216 | -16% | 0 | 0 | — |
case-16 | pass→pass | 15,373 | 13,005 | -15% | 1 | 1 | 0% | 2,304 | 1,916 | -17% | 0 | 0 | — |
case-17 | pass→pass | 16,908 | 4,645 | -73% | 1 | 1 | 0% | 1,822 | 1,356 | -26% | 0 | 0 | — |
case-18 | pass→pass | 11,703 | 9,519 | -19% | 1 | 1 | 0% | 908 | 1,316 | +45% | 0 | 0 | — |
case-19 | pass→pass | 19,932 | 19,313 | -3% | 1 | 1 | 0% | 2,609 | 3,900 | +49% | 0 | 0 | — |
case-20 | pass→pass | 36,610 | 35,942 | -2% | 1 | 1 | 0% | 5,034 | 6,827 | +36% | 0 | 0 | — |
case-21 | pass→pass | 11,475 | 20,123 | +75% | 1 | 1 | 0% | 2,399 | 3,335 | +39% | 0 | 0 | — |
case-22 | pass→pass | 14,506 | 8,079 | -44% | 1 | 1 | 0% | 1,500 | 1,052 | -30% | 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 +18 percentage points is the difference between those two pass rates over the 22 comparable cases.
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.