Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Deep claim scope analysis — decompose independent/dependent claims and assess protection scope breadth. Budget: 30 patent families, 30 claim parses, 20 web searches.
.claude/skills/yogsoth-ai-claim-analysis/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-04 | ✗→✓ | ▲ Improved | 41% | 0% |
| case-07 | ✗→✓ | ▲ Improved | -11% | 0% |
| case-08 | ✗→✓ | ▲ Improved | -24% | 0% |
| case-09 | ✗→✓ | ▲ Improved | 55% | 0% |
| case-10 | ✗→✓ | ▲ Improved | -24% | 0% |
Deep analysis of patent claim scope. Decomposes claim structure, maps element relationships, and assesses the breadth and strength of patent protection.
Provide detailed claim-level analysis for a set of target patents. Determines protection scope, identifies claim vulnerabilities, and maps the relationship between independent and dependent claims.
| Metric | Target | |--------|--------| | Patent families analyzed | 30 | | Claim parses | 30 | | Web searches completed | 20 |
| Metric | Target | Current | % | |--------|--------|---------|---| | Patent families analyzed | 30 | 0 | 0% | | Claim parses completed | 30 | 0 | 0% | | Web searches completed | 20 | 0 | 0% | | Independent claims parsed | — | 0 | — | | Claim elements extracted | — | 0 | — |
HARD-GATE: Cannot exit iteration loop until 80% of claim parses (24) and patent families (24) budget met.
| Tactic | When to Use | |--------|-------------| | claim-decomposition | Primary tactic — parse all claims and extract elements | | patent-family-tracing | Find family members with different claim scope | | classification-navigation | Understand technical context of claim elements |
| SOP | Role in This Strategy | |-----|----------------------| | claim-parsing | Core SOP — parse claim syntax and extract elements | | patent-query-formulation | Find related patents for scope comparison | | citation-network-analysis | Identify patents that cite or are cited by target | | legal-status-assessment | Confirm target patents are active | | quality-scoring | Assess claim quality (breadth, specificity, support) | | patent-synthesis | Produce claim analysis report |
legal-status-assessmentclaim-decomposition tactic for full structural analysismarkdown# Claim Analysis Report: [Patent Number(s)] ## Patent Overview [Title, assignee, filing date, status] ## Claim Structure ### Independent Claims | Claim # | Type | Element Count | Scope Assessment | |---------|------|---------------|------------------| ### Claim Tree [Dependency hierarchy visualization] ## Element-by-Element Analysis ### Claim 1 | Element | Technical Feature | Breadth | Potential Design-Around | |---------|-------------------|---------|------------------------| ## Scope Assessment - Broadest claim: [claim #, scope description] - Narrowest claim: [claim #, scope description] - Key limitations: [elements that narrow scope] ## Vulnerability Analysis - Prior art risks: [elements with known prior art] - Enablement concerns: [overly broad elements] - Design-around paths: [alternative implementations] ## Cross-Jurisdiction Comparison [Claim scope differences across patent family members]
<!-- 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 | | knowledge-acquisition-claim-decomposition | Independent/dependent claim parsing, element extraction, and feature mapping to technical domains | | 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 | | --- | --- | | citation-network-analysis | Build and analyze patent citation networks — main path analysis, PageRank, cluster detection | | claim-parsing | Patent claim syntax parsing — independent/dependent relationships and element extraction | | 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 | | 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 |
<!-- 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-18 | pass→pass | 12,707 | 9,255 | -27% | 1 | 1 | 0% | 2,034 | 2,653 | +30% | 0 | 0 | — |
case-01 | fail→fail | 39,112 | 7,517 | -81% | 1 | 1 | 0% | 6,199 | 1,450 | -77% | 0 | 0 | — |
case-02 | fail→fail | 28,122 | 35,412 | +26% | 1 | 1 | 0% | 4,606 | 1,591 | -65% | 0 | 0 | — |
case-03 | fail→fail | 27,194 | 10,924 | -60% | 1 | 1 | 0% | 4,553 | 1,666 | -63% | 0 | 0 | — |
case-04 | fail→pass | 27,780 | 2,636 | -91% | 1 | 1 | 0% | 1,015 | 1,435 | +41% | 0 | 0 | — |
case-05 | pass→pass | 10,184 | 2,346 | -77% | 1 | 1 | 0% | 1,627 | 1,504 | -8% | 0 | 0 | — |
case-06 | pass→pass | 12,889 | 5,063 | -61% | 1 | 1 | 0% | 2,116 | 1,843 | -13% | 0 | 0 | — |
case-17 | pass→pass | 17,740 | 9,665 | -46% | 1 | 1 | 0% | 2,584 | 2,564 | -1% | 0 | 0 | — |
case-07 | fail→pass | 9,889 | 2,689 | -73% | 1 | 1 | 0% | 1,809 | 1,614 | -11% | 0 | 0 | — |
case-08 | fail→pass | 12,704 | 2,963 | -77% | 1 | 1 | 0% | 2,110 | 1,605 | -24% | 0 | 0 | — |
case-09 | fail→pass | 8,082 | 5,460 | -32% | 1 | 1 | 0% | 1,277 | 1,978 | +55% | 0 | 0 | — |
case-10 | fail→pass | 13,257 | 2,984 | -77% | 1 | 1 | 0% | 2,076 | 1,568 | -24% | 0 | 0 | — |
case-11 | fail→pass | 12,416 | 2,849 | -77% | 1 | 1 | 0% | 2,305 | 1,563 | -32% | 0 | 0 | — |
case-12 | fail→pass | 7,576 | 2,000 | -74% | 1 | 1 | 0% | 1,022 | 1,344 | +32% | 0 | 0 | — |
case-13 | pass→pass | 17,728 | 12,297 | -31% | 1 | 1 | 0% | 2,810 | 2,868 | +2% | 0 | 0 | — |
case-14 | pass→pass | 16,503 | 12,084 | -27% | 1 | 1 | 0% | 2,625 | 2,826 | +8% | 0 | 0 | — |
case-15 | pass→pass | 11,343 | 8,956 | -21% | 1 | 1 | 0% | 1,813 | 2,426 | +34% | 0 | 0 | — |
case-16 | fail→pass | 18,191 | 10,783 | -41% | 1 | 1 | 0% | 2,729 | 2,745 | +1% | 0 | 0 | — |
case-19 | fail→fail | 7,814 | 5,254 | -33% | 1 | 1 | 0% | 1,212 | 2,006 | +66% | 0 | 0 | — |
case-20 | pass→pass | 7,077 | 23,552 | +233% | 1 | 1 | 0% | 1,183 | 4,406 | +272% | 0 | 0 | — |
case-21 | pass→pass | 13,730 | 27,913 | +103% | 1 | 1 | 0% | 2,888 | 6,074 | +110% | 0 | 0 | — |
case-22 | pass→pass | 12,797 | 15,929 | +24% | 1 | 1 | 0% | 2,349 | 3,810 | +62% | 0 | 0 | — |
case-23 | pass→pass | 20,627 | 26,741 | +30% | 1 | 1 | 0% | 3,653 | 5,646 | +55% | 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. 23 cases were attempted, and 19 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 +35 percentage points is the difference between those two pass rates over the 19 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.