使用方法
- 复制安装命令,在自己的智能体环境安装前先检查源码包。
- 从原始文档核对所需工具、凭据和支持的宿主。
- 说明输入材料与期望交付物,并对照来源检查输出。
npx skills add https://github.com/yogsoth-ai/de-anthropocentric-research-engine --skill parse-patent-claim开始前需了解
- 安装不包含模型订阅、API 额度或访问文件的权限。
- 仓库星数和安装量反映活跃情况,不等于独立质量或安全测试。
按源仓库内容呈现,保留标题、案例、代码、表格、链接以及原文引用的演示图片。
parse-patent-claim
Purpose
Parse independent and dependent claim structure, extract elements, and map claim features to technical functions or domains.
Input contract
required: [claim_text, claim_set]
optional: [technical_ontology, family_context, jurisdiction]
constraints: [element boundaries, dependency links, and source spans must be retained]Procedure
- Identify independent claims and dependency relations.
- Segment each claim into required elements, limitations, and functional language.
- Map elements to technical functions and preserve alternative interpretations.
- Emit a source-linked claim-element graph.
If claim elements, dependencies, and limiting language are parsed, consider assess-patent-claim-scope as the next tactic.
Output contract
produces: [claim_element_map, dependency_graph, function_map, interpretation_alternatives]
delta_fields: [findings, evidence_updates, uncertainties, open_questions]Quality gates
- Every element points to a claim span.
- Dependency and optionality are not conflated.
Failure and counterexamples
Do not paraphrase away limiting language or treat a dependent claim as independent coverage.
Provenance map
resolved: claim-parsingresolved: knowledge-acquisition-claim-decomposition


