Menu
生成符合中国专利法规范的专利交底书。接收 Markdown 格式的技术描述作为输入，通过多阶段流水线（标题生成、专利检索、发明目的撰写、发明内容撰写、图表生成、报告合并）自动生成完整的专利交底书。
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Based on SOC occupation classification
Reviews code diffs, PRs, issue queues, release readiness, commits, pushes, publishing, and project audits. Use when users ask in any language for code review, issue or PR triage, release gates, publishing follow-through, or project audits. Not for debugging root causes or prose review.
Produces distinctive, production-grade UI for pages, components, visual interfaces, typography, and screenshot-driven polish. Use when users ask in any language for UI, page, component, frontend, typography, screenshot-grounded visual polish, or complaints that a screen looks unclear, ugly, inconsistent, or visually wrong. Not for backend logic or data pipelines.
Use when the user requests diagrams, flowcharts, architecture diagrams, ER diagrams, UML / sequence / class diagrams, network topology, ML/DL model figures (Transformer/CNN/LSTM), mind maps, or any visualization. Also use proactively when explaining systems with 3+ components, complex data flows, or relationships that benefit from visual representation. Best suited when the diagram needs custom styling, rich shape vocabulary, swimlanes, or exportable images (PNG/SVG/PDF/JPG). Generates .drawio XML and exports locally via the native draw.io desktop CLI.
Runs a budget-aware agent-assisted engineering health audit for instruction/config drift, hooks/MCP, verifier surfaces, and AI maintainability. Use when users ask in any language to audit Claude, Codex, Pi, agent instructions, MCP or hooks, verifier coverage, or AI-maintainability drift. Not for debugging application code or reviewing PRs.
Finds root cause before applying fixes for errors, crashes, regressions, failing tests, broken behavior, and screenshot-reported defects. Use when users report in any language errors, crashes, broken behavior, regressions, failing tests, screenshot evidence, or something that used to work and now fails. Not for code review or new features.
Runs a six-phase research workflow that turns unfamiliar domains, source bundles, or collected material into publish-ready output. Use when users ask in any language to research, study, deep-dive, compile sources, synthesize unfamiliar material, or turn a source bundle into a coherent reference. Not for quick lookups or single-file reads.
| name | patent-draft-agent |
| description | 生成符合中国专利法规范的专利交底书。接收 Markdown 格式的技术描述作为输入，通过多阶段流水线（标题生成、专利检索、发明目的撰写、发明内容撰写、图表生成、报告合并）自动生成完整的专利交底书。 |
你负责协调多个子代理，按阶段产出完整、可提交审阅的专利交底书。
API Key 要求：
SERPAPI_API_KEY
google-patents-mcp.search_patents，这是 Google Patents 检索的必需凭据EXA_API_KEY
exa.web_search_exa / exa.crawling_exa开始前先运行：
bash <SKILL_DIR>/scripts/init.sh
如果失败，停止执行，并明确缺失的环境变量或命令。
从输出中记录 SKILL_DIR=... 的绝对路径。后续所有 skill 内文件都基于该路径引用。
本 skill 的检索兼容两层能力：
mcporter CLI 执行固定格式：
mcporter call <server>.<tool> --config <SKILL_DIR>/references/mcporter.json <key=value>...
mcporter 可用服务：
| 服务 | 工具 | 典型用途 |
|---|---|---|
google-patents-mcp | search_patents | Google Patents 专利检索 |
exa | web_search_exa | 可选的网页 / 论文 / 产品语义检索 |
exa | crawling_exa | 可选的重点页面正文抓取 |
调用规则：
search_patents 至少传 query 与 num_resultsweb_search_exa 至少传 query 与 numResultscrawling_exa 传 JSON 数组格式的 urls示例：
mcporter call google-patents-mcp.search_patents \
--config <SKILL_DIR>/references/mcporter.json \
query="大规模文件分发" \
num_results=8
mcporter call exa.web_search_exa \
--config <SKILL_DIR>/references/mcporter.json \
query="中国专利 文件分发系统" \
numResults=5
收到请求后创建工作目录：
uuid=$(python3 -c "import uuid;print(uuid.uuid4())")
WORK_DIR=output/temp_${uuid}
mkdir -p ${WORK_DIR}/{00_input,01_title,02_search,03_sections,04_final,metadata}
如果用户提供 UUID，则复用该 UUID。
将用户输入保存为 ${WORK_DIR}/00_input/input.md。
严格按以下 6 个阶段执行。每个阶段结束后先验收文件，再进入下一阶段。
对子代理的统一要求：
mcporter.json 的绝对路径。Prompt：<SKILL_DIR>/references/prompts/title-generator.md
输入：
${WORK_DIR}/00_input/input.md输出：
${WORK_DIR}/01_title/title.txt验收：
Prompt：<SKILL_DIR>/references/prompts/patent-searcher.md
输入：
${WORK_DIR}/00_input/input.md${WORK_DIR}/01_title/title.txt输出：
${WORK_DIR}/02_search/search-report.md${WORK_DIR}/02_search/similar-patents-reference.md验收：
search-report.md 包含检索关键词、相关专利、区别点总结similar-patents-reference.md 包含术语规范和写作建议Prompt：<SKILL_DIR>/references/prompts/purpose-writer.md
输入：
${WORK_DIR}/00_input/input.md${WORK_DIR}/01_title/title.txt${WORK_DIR}/02_search/similar-patents-reference.md输出：
${WORK_DIR}/03_sections/purpose-section.md验收：
Prompt：<SKILL_DIR>/references/prompts/content-writer.md
输入：
${WORK_DIR}/00_input/input.md${WORK_DIR}/01_title/title.txt${WORK_DIR}/02_search/similar-patents-reference.md${WORK_DIR}/03_sections/purpose-section.md输出：
${WORK_DIR}/03_sections/content-section.md验收：
Prompt：<SKILL_DIR>/references/prompts/diagram-generator.md
输入：
${WORK_DIR}/03_sections/content-section.md输出：
${WORK_DIR}/03_sections/diagrams.md验收：
Prompt：<SKILL_DIR>/references/prompts/report-merger.md
输入：
${WORK_DIR}/01_title/title.txt${WORK_DIR}/02_search/search-report.md${WORK_DIR}/03_sections/purpose-section.md${WORK_DIR}/03_sections/content-section.md${WORK_DIR}/03_sections/diagrams.md输出：
${WORK_DIR}/04_final/output.md${WORK_DIR}/metadata/quality-check.json验收：
output.md 存在quality-check.json 存在对子代理不要只给“请处理”这类模糊指令，必须给出完整执行框架。推荐模板：
你现在执行 skill 子任务：<agent-name>
先完整阅读以下 prompt 文件，再开始工作：
<PROMPT_FILE_ABS_PATH>
执行约束：
- 工作目录：<WORK_DIR_ABS_PATH>
- 仅允许读取以下输入文件：
- <INPUT_FILE_1_ABS_PATH>
- <INPUT_FILE_2_ABS_PATH>
- 仅允许写入以下输出文件：
- <OUTPUT_FILE_1_ABS_PATH>
- <OUTPUT_FILE_2_ABS_PATH>
- 除非任务明确要求，不得修改其他文件
如任务涉及检索：
- Google Patents 必须使用 mcporter：
- mcporter 配置：<SKILL_DIR>/references/mcporter.json
- 命令格式：mcporter call google-patents-mcp.search_patents --config <CONFIG_ABS_PATH> key=value ...
- 产品 / 论文 / 网页资料优先使用 Exa；若 Exa 不可用，则回退到 coding agent 内置搜索/浏览能力
- 先执行检索，再整理成 Markdown 输出
完成条件：
- 所有指定输出文件都已写入
- 内容满足 prompt 中的结构和字数要求
- 最终回复中仅总结完成情况与输出路径
允许：
禁止：