Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Patent landscape full-scan — maps technology domain via assignee ranking, IPC/CPC classification, filing trends. Budget: 200 patent families, 0 claim parses, 80 web searches.
.claude/skills/yogsoth-ai-landscape-survey/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-05 | ✗→✓ | ▲ Improved | -28% | 0% |
| case-04 | ✗→✓ | ▲ Improved | 22% | 0% |
| case-06 | ✗→✓ | ▲ Improved | -35% | 0% |
| case-13 | ✗→✓ | ▲ Improved | 51% | 0% |
| case-14 | ✗→✓ | ▲ Improved | 44% | 0% |
Full-scan patent landscape analysis. Maps the technology domain by identifying key assignees, classifying by IPC/CPC taxonomy, and analyzing filing trends over time.
Produce a comprehensive patent landscape map for a given technology domain, answering: Who are the key players? What are the main technology branches? How is filing activity trending?
| Metric | Target | |--------|--------| | Patent families analyzed | 200 | | Claim parses | 0 | | Web searches completed | 80 |
| Metric | Target | Current | % | |--------|--------|---------|---| | Patent families analyzed | 200 | 0 | 0% | | Web searches completed | 80 | 0 | 0% | | Assignees identified | — | 0 | — | | IPC/CPC classes mapped | — | 0 | — |
HARD-GATE: Cannot exit iteration loop until 80% of patent families (160) and web searches (64) budget met.
| Tactic | When to Use | |--------|-------------| | patent-family-tracing | Expand from seed patents to full family coverage | | classification-navigation | Drill into IPC/CPC hierarchy to find all relevant subclasses |
| SOP | Role in This Strategy | |-----|----------------------| | patent-query-formulation | Generate initial search queries for the domain | | patent-categorization | Classify discovered patents into technology sub-domains | | assignee-normalization | Standardize assignee names across patent offices | | trend-analysis | Analyze filing volume time-series and lifecycle curves | | quality-scoring | Rank patent families by quality metrics | | saturation-detection | Check if landscape scan has converged | | patent-synthesis | Produce final landscape report |
patent-query-formulation to generate keyword + IPC/CPC + assignee search strategiespatent-family-tracing tactic to expand from top resultsclassification-navigation tactic to find related technology branchespatent-categorizationassignee-normalizationtrend-analysis on the full datasetsaturation-detection to confirm convergencepatent-synthesismarkdown# Patent Landscape Report: [Technology Domain] ## Executive Summary [Key findings in 3-5 bullets] ## Top Assignees (ranked by family count) | Rank | Assignee | Families | Key IPC Classes | |------|----------|----------|-----------------| ## Technology Classification Map [IPC/CPC hierarchy with patent counts per node] ## Filing Trends [Time-series data: year, total filings, top assignee filings] ## Technology Clusters [Grouped by application domain / technology branch] ## Appendix: Full Patent Family List [Structured table of all 200 families]
<!-- BEGIN available-tables (generated) -->
Optional, no fixed order; the final leaf is always a sop.
| Tactic | When to use | | --- | --- | | classification-navigation | IPC/CPC hierarchy drill-down and lateral expansion for patent discovery | | patent-family-tracing | Forward/backward patent citation and priority tracing until saturation |
Optional, no fixed order; the final leaf is always a sop.
| SOP | When to use | | --- | --- | | assignee-normalization | Standardize assignee names and identify corporate group affiliations across patent offices | | 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. | | patent-categorization | Classify patents by tech subdomain, application scenario, and value chain position | | patent-query-formulation | Construct keyword + IPC/CPC + assignee combination search strategies for patent databases | | patent-synthesis | Produce final structured patent intelligence report from all analysis results | | quality-scoring | Multi-dimensional patent quality assessment — forward citations, family size, claim count, geographic breadth | | trend-analysis | Patent filing volume time-series, technology lifecycle stage, and S-curve analysis |
<!-- 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-05 | fail→pass | 21,133 | 8,608 | -59% | 1 | 1 | 0% | 2,048 | 1,478 | -28% | 0 | 0 | — |
case-01 | fail→fail | 55,354 | 65,649 | +19% | 1 | 1 | 0% | 8,286 | 2,049 | -75% | 0 | 0 | — |
case-02 | fail→fail | 63,965 | 89,287 | +40% | 1 | 1 | 0% | 8,291 | 1,711 | -79% | 0 | 0 | — |
case-03 | fail→fail | 49,126 | 64,728 | +32% | 1 | 1 | 0% | 8,280 | 9,327 | +13% | 0 | 0 | — |
case-04 | fail→pass | 23,098 | 19,170 | -17% | 1 | 1 | 0% | 2,557 | 3,111 | +22% | 0 | 0 | — |
case-10 | pass→fail | 22,628 | 42,435 | +88% | 1 | 1 | 0% | 2,424 | 7,156 | +195% | 0 | 0 | — |
case-06 | fail→pass | 14,122 | 7,393 | -48% | 1 | 1 | 0% | 2,068 | 1,352 | -35% | 0 | 0 | — |
case-07 | fail→fail | 13,594 | 9,275 | -32% | 1 | 1 | 0% | 1,075 | 1,553 | +44% | 0 | 0 | — |
case-08 | fail→fail | 21,458 | 18,289 | -15% | 1 | 1 | 0% | 2,085 | 3,135 | +50% | 0 | 0 | — |
case-09 | fail→fail | 15,944 | 12,887 | -19% | 1 | 1 | 0% | 1,967 | 2,303 | +17% | 0 | 0 | — |
case-11 | pass→fail | 10,227 | 47,275 | +362% | 1 | 1 | 0% | 1,834 | 9,985 | +444% | 0 | 0 | — |
case-12 | pass→pass | 17,362 | 33,946 | +96% | 1 | 1 | 0% | 2,716 | 6,830 | +151% | 0 | 0 | — |
case-13 | fail→pass | 27,924 | 7,855 | -72% | 1 | 1 | 0% | 961 | 1,450 | +51% | 0 | 0 | — |
case-14 | fail→pass | 34,579 | 8,069 | -77% | 1 | 1 | 0% | 1,003 | 1,449 | +44% | 0 | 0 | — |
case-15 | fail→pass | 11,862 | 3,129 | -74% | 1 | 1 | 0% | 1,151 | 1,354 | +18% | 0 | 0 | — |
case-16 | pass→pass | 15,000 | 15,124 | +1% | 1 | 1 | 0% | 2,350 | 2,424 | +3% | 0 | 0 | — |
case-17 | pass→pass | 19,903 | 2,540 | -87% | 1 | 1 | 0% | 1,742 | 1,398 | -20% | 0 | 0 | — |
case-18 | fail→pass | 13,404 | 3,628 | -73% | 1 | 1 | 0% | 1,950 | 1,372 | -30% | 0 | 0 | — |
case-19 | fail→fail | 10,398 | 3,367 | -68% | 1 | 1 | 0% | 1,594 | 1,546 | -3% | 0 | 0 | — |
case-20 | pass→pass | 21,466 | 3,186 | -85% | 1 | 1 | 0% | 1,631 | 1,513 | -7% | 0 | 0 | — |
case-21 | pass→pass | 4,930 | 3,132 | -36% | 1 | 1 | 0% | 295 | 1,262 | +328% | 0 | 0 | — |
case-22 | fail→fail | 4,412 | 5,487 | +24% | 1 | 1 | 0% | 589 | 1,335 | +127% | 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, and 18 counted toward the lift figure. The other 4 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of +23 percentage points is the difference between those two pass rates over the 18 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.