nature-paper-to-patent

Convert scientific papers, theses, technical reports, source code, figures, inventor notes, or research manuscripts into evidence-grounded Chinese invention patent drafts and attorney-facing technical disclosure materials. Use when an AI agent must mine patent points, draft or revise a Chinese technical disclosure (技术交底书), run prior-art comparison, convert Office project materials, map every claimed feature to source evidence, preserve core formulas as editable Office Math, generate claim-aligned flowcharts and methodology figures, compare a paper with an existing patent, audit support and consistency, or deliver Chinese DOCX patent/disclosure files.

Quality

88%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Quality

Content

77%

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

A well-structured, concise router with strong workflow sequencing and validation gates, plus real executable validation/build commands. Its main weakness is that the manifest-based routing core (manifest.yaml and the static/ fragments it points to) is not present in the bundle, which makes the load entry point and detailed gate criteria non-executable as shipped.

Suggestions

Ship manifest.yaml and the static/core/ and static/fragments/ files the body references, or inline the always_load list and stage-gate criteria so the skill is self-contained and the load step actually executes.

Add a fallback in section 1 for when manifest.yaml is unavailable (e.g., proceed with the always_load defaults listed inline) so the entry point is not a dead end.

Confirm every path cited in the body (e.g., static/core/output-contract.md, static/fragments/task/technical-disclosure.md) exists in the bundle or convert dangling references into inline guidance.

Dimension	Reasoning	Score
Conciseness	Lean directive prose that assumes Claude's competence — no preamble explaining what a patent or Office Math is — and every line instructs (ID schemes, support states, stage gates, output languages). Not a 2 because there is no unnecessary explanation or padding to tighten.	3 / 3
Actionability	Gives concrete, copy-paste-ready commands ('python scripts/validate_patent_draft.py draft.json', 'python scripts/build_patent_package.py draft.json --output-dir outputs --prefix patent') and specific schemas (P001/E001/F001/C001; explicit/inherent/needs-confirmation/unsupported), but the primary load entry point ('Read manifest.yaml, then read every file under always_load') and the static/core/workflow.md, static/fragments/task/* targets are absent from the bundle, leaving the load steps non-executable as written.	2 / 3
Workflow Clarity	Five ordered stages (Load → Preserve grounding → Draft through stage gates → Produce formal documents → Validate) with explicit gates ('Do not move to formal claims until the source map, terminology ledger, inventories, evidence ledger, and invention concept pass their gates') and a validate→resolve-errors→deliver feedback loop. Not a 2 because sequencing and validation checkpoints are explicit in the body.	3 / 3
Progressive Disclosure	Has a clear 5-section overview with one-level-deep references to real files (references/draft-schema.md, references/disclosure/disclosure_self_check.md, scripts/...), but the routing linchpin — manifest.yaml and the entire static/ tree (workflow.md, fragments, output-contract.md) — is referenced yet missing from the bundle, breaking the router→manifest→fragment navigation. Not a 3 because of these dangling entry-point paths; not a 1 because genuine structure and several valid references are present.	2 / 3
	Total	10 / 12 Passed

Description

100%

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

An excellent description: concrete capabilities, an explicit 'Use when' trigger enumerating many natural scenarios, and a sharply defined Chinese-patent niche that minimizes conflict risk. No vague fluff or over-claims.

Dimension	Reasoning	Score
Specificity	Lists many concrete actions — 'mine patent points', 'run prior-art comparison', 'map every claimed feature to source evidence', 'preserve core formulas as editable Office Math', 'generate claim-aligned flowcharts', 'audit support and consistency' — matching the 'multiple specific concrete actions' anchor.	3 / 3
Completeness	Explicitly answers both: what ('Convert ... into evidence-grounded Chinese invention patent drafts and attorney-facing technical disclosure materials') and when ('Use when an AI agent must ...'), with an explicit trigger clause.	3 / 3
Trigger Term Quality	Covers natural phrasing a researcher would use — 'scientific papers, theses, technical reports, source code, figures, inventor notes', 'draft or revise a Chinese technical disclosure (技术交底书)', 'compare a paper with an existing patent', 'deliver Chinese DOCX patent/disclosure files' — with good breadth and common variations.	3 / 3
Distinctiveness Conflict Risk	Occupies a clear, narrow niche (Chinese invention patent drafting / 技术交底书) with distinct triggers unlikely to fire for unrelated skills.	3 / 3
	Total	12 / 12 Passed

Validation

93%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
referenced_paths_exist	Referenced path issues: 2 deeper-than-1-level	Warning

	Total	15 / 16 Passed

Repository: Yuan1z0825/nature-skills
Path: skills/nature-paper-to-patent/SKILL.md
Commit: 703c150

Reviewed: 4 days ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.