patent-write

基于用户提供的技术方案、草稿或参考专利，生成、改写与统稿中文发明专利的题目、摘要、背景技术、发明内容、权利要求、附图说明和具体实施方式。用于专利撰写、专利摘要、权利要求、说明书、专利润色、专利仿写、参考专利蒸馏等场景。

Quality

81%

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

71%Weight 40%Scale 1-5

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

Well-structured, concise, and actionable body with a clear routing table and sequenced workflow. The main weakness is progressive disclosure: the body points to 11 reference files that are not present in the bundle.

Suggestions

Ship the referenced reference/*.md bundle files (full-patent.md, title.md, abstract.md, background.md, invention-content.md, claims.md, figures-desc.md, figures-drawio.md, implementation.md, review.md, distill.md) so the routing table links resolve.

Add an explicit validation/verify checkpoint in the workflow (e.g., a concrete 自检 step with pass/fail criteria) rather than leaving review implicit via the reference.

Trim the slight overlap between 通用原则 and 从参考专利蒸馏出的高频规律 (e.g., 术语统一 and 先宽后窄 appear in both) to tighten conciseness.

Dimension	Reasoning	Score
Conciseness	The body is lean, uses a routing table and bullet lists, and never explains concepts Claude already knows; minor redundancy between 通用原则 and 从参考专利蒸馏出的高频规律 keeps it just below a 5.	4 / 5
Actionability	A keyword-to-file routing table, a concrete input-collection checklist, and an ordered full-patent procedure give mostly executable guidance, though the actual writing rules live in referenced files.	4 / 5
Workflow Clarity	Steps 1–3 are clearly sequenced with an ordered full-patent sub-procedure and a closing 统稿/一致性检查 review checkpoint plus an insufficiency fallback; validation is somewhat implicit rather than an explicit verify step.	4 / 5
Progressive Disclosure	The in-body structure is well-signaled with one-level-deep [reference/*.md] links in a routing table, but none of the referenced bundle files actually exist (no reference/ directory), so the promised navigation breaks.	3 / 5
	Total	15 / 20 Passed

Description

92%Weight 40%Scale 1-5

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

A strong description: third-person voice, concrete actions across named patent sections, and an explicit 'when' trigger clause with natural terms. Only minor room to add a few more trigger synonyms.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions (生成、改写、统稿) applied to eight named patent sections (题目、摘要、背景技术、发明内容、权利要求、附图说明、具体实施方式), giving comprehensive coverage rather than vague language.	5 / 5
Completeness	It explicitly states both the 'what' (generates, rewrites, integrates specific patent sections) and the 'when' via the 用于…等场景 trigger clause with concrete trigger phrases.	5 / 5
Trigger Term Quality	The 用于…场景 clause surfaces many natural user phrases (专利撰写、专利摘要、权利要求、说明书、专利润色、专利仿写、参考专利蒸馏), but coverage is strong rather than exhaustive — a few common synonyms are absent.	4 / 5
Distinctiveness Conflict Risk	The niche is tightly scoped to 中文发明专利撰写 with patent-specific triggers, making conflict with unrelated skills minimal.	5 / 5
	Total	19 / 20 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
relative_links	Relative link issues: 11 missing	Warning

	Total	15 / 16 Passed

Repository: xstongxue/best-skills
Path: skills/patent-write/SKILL.md
Commit: 88a38bb

Reviewed: ~1 hour 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.