FreePublished on GitHubEnglish overview

nature-paper-to-patent

Turn research papers and inventor materials into Chinese invention patent drafts or technical disclosures. Link claimed features to source evidence, align terminology across documents, and produce checked DOCX materials for inventor and patent-professional review.

OriginalShow the original description

Turn research papers or inventor materials into evidence-grounded Chinese invention patent drafts and technical disclosures. Use for 技术交底书、专利撰写、现有技术对比 and Chinese DOCX patent packages; not general manuscript writing.

Yuan1z0825/nature-skills

Run score Not tested

See how to install

Good for

Draft Chinese patent documents from a research paper
Prepare or revise a Chinese technical disclosure
Trace patent claim features to source evidence
Check consistency across claims, figures, and descriptions

Overview

What it does

Creates Chinese invention patent drafts or technical disclosures (技术交底书) from research papers and materials supplied by inventors. It assigns identifiers to source text, equations, figures, and code, then maps each technical feature in a claim to its supporting evidence. For a formal application package, it develops the claims, specification, figures, embodiments, and abstract in a consistent order and checks terminology and procedural consistency. Technical disclosures are delivered as Markdown and DOCX. For algorithmic inventions, formulas in the DOCX are editable Office Math elements.

When it helps

Use it to organize research results into a Chinese patent draft or prepare review materials for inventors and patent professionals. It can also revise an existing technical disclosure or compare a paper with a patent draft.

Example requests

“Create a Chinese patent draft from this paper and link each claim to its supporting evidence.”
“Revise this technical disclosure and provide a DOCX version.”
“Check how the claims map to the source and whether the figures match the specification.”

Notes

Formal deliverables are in Chinese. Document generation and validation use Python scripts, and diagrams use Mermaid. Features unsupported by the source are excluded from formal claims, and unknown points are marked. The output is a draft for review by inventors and patent professionals; it does not determine patentability or infringement and does not guarantee filing.

This overview was written by AI (OpenAI Codex) from the public SKILL.md. Check the original below for the exact behavior.

Document check

AI checked the SKILL.md for operations that need care and for setup requirements. The score comes only from actually running the skill (hands-on test).

SafetyAre sending data, deleting files, and running commands disclosed?: BPartly lackingIt runs Python validation and build scripts, but gives no explicit pre-run confirmation step.
Easy setupDoes it need few tools or API keys, and are they documented?: BPartly lackingPython and Mermaid are required, but installation steps and environment details are not given.

Not counted in the score. Rated by AI (OpenAI Codex) as of Oct 24, 2026. Treat it as a guide and read the original before installing.

Disagree? Request a re-evaluation

Files included (69)

SKILL.md4.5 KB
agents/openai.yaml248 B
evals/evals.json1.5 KB
manifest.yaml3.1 KB
README_EN.md3.3 KB
README.md2.9 KB
references/cn-patent-drafting-guide.md8.0 KB
references/corpus-derived-patterns.md3.5 KB
references/corpus-pair-audit.md2.9 KB
references/disclosure/correction_handler.md2.8 KB
references/disclosure/disclosure_builder.md12.1 KB
references/disclosure/disclosure_preview.md917 B
references/disclosure/disclosure_self_check.md5.9 KB
references/disclosure/intake.md943 B
references/disclosure/iteration_context.md4.3 KB
references/disclosure/merger.md3.4 KB
references/disclosure/patent_points_analyzer.md951 B
references/disclosure/patent-disclosure-skill-MIT-LICENSE.txt1.0 KB
references/disclosure/prior_art_search.md10.4 KB
references/disclosure/project_scan.md3.6 KB
references/disclosure/template_reference.md9.2 KB
references/disclosure/tooling.md16.2 KB
references/draft-schema.md9.0 KB
references/patent-figure-guide.md2.9 KB
requirements.txt121 B
scripts/audit_claims.py6.4 KB
scripts/build_patent_package.py6.5 KB
scripts/disclosure/cnipa_epub_crawler.py10.5 KB
scripts/disclosure/cnipa_epub_parse.py8.6 KB
scripts/disclosure/cnipa_epub_search.py5.8 KB
scripts/disclosure/docx_to_md.py3.7 KB
scripts/disclosure/iteration_dialog_log.py3.2 KB
scripts/disclosure/math_render.py12.7 KB
scripts/disclosure/md_to_docx.py32.2 KB
scripts/disclosure/mermaid_render.py15.2 KB
scripts/disclosure/package-lock.json84.4 KB
scripts/disclosure/package.json279 B
scripts/disclosure/pptx_to_md.py4.9 KB
scripts/disclosure/requirements-cnipa.txt172 B
scripts/extract_pdf_text.py2.3 KB
scripts/init_patent_project.py3.1 KB
scripts/math_to_omml.py4.8 KB
scripts/render_flowchart_svg.py14.3 KB
scripts/render_patent_docx.py10.3 KB
scripts/validate_patent_draft.py14.6 KB
static/core/output-contract.md2.0 KB
static/core/principles.md1016 B
static/core/workflow.md3.8 KB
static/fragments/invention/algorithm-software.md366 B
static/fragments/invention/apparatus-system.md288 B
static/fragments/invention/mixed.md255 B
static/fragments/invention/process-material.md278 B
static/fragments/source/mixed-project.md369 B
static/fragments/source/pasted-text.md316 B
static/fragments/source/pdf-text.md290 B
static/fragments/source/scanned-pdf.md272 B
static/fragments/task/claim-set.md249 B
static/fragments/task/disclosure-analysis.md236 B
static/fragments/task/disclosure-iteration.md1005 B
static/fragments/task/full-draft.md247 B
static/fragments/task/paper-patent-audit.md280 B
static/fragments/task/technical-disclosure.md1.5 KB
tests/test_audit_claims.py1.6 KB
tests/test_inline_math_paren.py1.7 KB
tests/test_math_to_omml_superscript.py2.2 KB
tests/test_md_to_docx_source_overwrite.py1.4 KB
tests/test_mermaid_render.py2.3 KB
tests/test_validation_figure_node_ids.py614 B
tests/test_validation.py6.0 KB

SKILL.md (original)

Review the instructions the agent will receive before you install.

Paper to Chinese Patent

Use this file as the router for the patent-drafting workflow. Do not draft the application directly from the paper abstract or contribution list.

1. Load the workflow

Read manifest.yaml, then read every file under always_load.

Detect these axes from the user's files and request:

source_format: selectable PDF, scanned PDF, pasted text, or mixed project;
task_mode: full draft, claim set, disclosure analysis, technical disclosure, disclosure iteration, or paper-patent audit;
invention_type: algorithm/software, apparatus/system, process/material, or mixed.

State the detected values in one short line. Load only the matching fragments declared in the manifest. Load detailed references only when their condition applies.

2. Preserve source grounding

Create stable source IDs before drafting:

P001... for paper text blocks;
E001... for equations;
F001... for source figures;
C001... for source-code or supplementary evidence.

Every material feature in a formal claim must map to one or more source IDs. Use only explicit, inherent, needs-confirmation, or unsupported as support states. Exclude unsupported features from formal claims.

Never infer inventorship, ownership, unpublished implementation details, publication dates, prior-art conclusions, or legal sufficiency. Use [TO CONFIRM: specific question] outside formal claims when facts are missing.

3. Draft through stage gates

For full-draft, claim-set, disclosure-analysis, and paper-patent-audit, complete the stages in static/core/workflow.md in order. Persist the intermediate artifacts specified there. Do not move to formal claims until the source map, terminology ledger, inventories, evidence ledger, and invention concept pass their gates.

For technical-disclosure, follow the ordered prompt references in static/fragments/task/technical-disclosure.md. For disclosure-iteration, follow static/fragments/task/disclosure-iteration.md and preserve the prior draft instead of restarting the formal application workflow.

For a full application, draft claims first, then align the specification, figures, embodiments, and abstract to the claim terminology and step order.

4. Produce Chinese formal documents

Agent-facing analysis may use the user's preferred language. Produce formal Chinese patent deliverables in Chinese when the task is a formal application package:

权利要求书;
说明书;
说明书摘要;
摘要附图;
figure labels and descriptions.

For technical-disclosure and disclosure-iteration, produce the Chinese technical disclosure (技术交底书) as timestamped Markdown plus matching DOCX, with Mermaid system/process diagrams rendered through scripts/disclosure/.

For algorithmic inventions, retain source-supported core formulas, define every symbol, explain each formula's technical operation, and render formulas as native editable Office Math in DOCX. Do not use plain LaTeX strings as the visible formula.

Generate the main flowchart from the ordered steps of the principal method claim. Its final node must name the concrete domain output, such as a defect detection result, target pose, state estimate, or control instruction. Reuse the same main figure as the abstract figure and a specification figure.

5. Validate before delivery

For formal application packages, populate the structured draft described in references/draft-schema.md, then run:

python scripts/validate_patent_draft.py draft.json
python scripts/build_patent_package.py draft.json --output-dir outputs --prefix patent

Resolve all validation ERROR findings. Review every WARNING against the source. Label the result incomplete draft when a required quality threshold in static/core/output-contract.md is not met.

For technical disclosures, run the internal checks in references/disclosure/disclosure_self_check.md, render Mermaid/Word outputs with scripts/disclosure/mermaid_render.py, and resolve formula, parameter, prior-art URL, and chapter-consistency issues before delivery.

The generated package is a drafting aid for inventor and patent-professional review, not a patentability opinion, infringement opinion, or filing guarantee.

Reviews

No reviews yet. Tell others what you thought.

Comments

No login needed. Share how it worked for you, ask a question, or note where it did not run.

No comments yet. Be the first to write one.

More skills from this repository

nature-academic-search

FreeSummary

Search academic literature across sources, verify references, and convert citation files. Build PubMed queries using MeSH, its subject vocabulary, or audit citations to distinguish independent citations and document who cited a paper and how.

Search academic literature across multiple sources
Verify references extracted from a document
Build a MeSH-based PubMed search strategy