CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

patent-search

中国专利公布公告著录检索：发明人、申请人、分类号、名称、摘要；也可从单图或权利要求生成名称/摘要关键字后查询。

Quality

79%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./skills/patent-search/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

78%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.

The body is exceptionally lean and well-sectioned (用途/何时用/输入/步骤/护栏/产出物) with concrete, copy-paste commands, explicit input rules, and guardrails. Its weakness is bundle integrity: every prompt/tool path it instructs Claude to Read is missing from the provided bundle, and the one existing reference file is orphaned from the body.

Suggestions

Ship the referenced files (prompts/patent_search.md, prompts/derived_query.md, prompts/covers_rank.md, tools/cnipa_search.py, tools/emit_search_report.py) in the bundle — the body's first instruction is to Read prompts/patent_search.md, which does not exist here, so the skill cannot execute as bundled.

Link references/patent_type_search.yaml from the body (e.g., from 输入 or 步骤 when describing --type) so the type-to-checkbox mapping table is discoverable instead of orphaned.

Inline one concrete example of the per-field query syntax (e.g., how 发明人 vs 摘要 fields differ) so the core command is executable without the missing prompt file.

DimensionReasoningScore

Conciseness

The ~40-line body is lean with zero concept explanations — every line instructs ("先 Read prompts/patent_search.md", one-line guardrails like "结果不得冒充交底查新"), fully assuming Claude's competence. Matches the anchor "every token earns its place".

5 / 5

Actionability

Four concrete copy-paste bash commands with realistic flags (--inventor, --class B01J20, --type design --derived-from image --type-inferred, --complete) plus machine-readable prefixes and output paths. Not 5 because the per-field query syntax is deferred ("填表语法跟公布站高级查询说明走") to prompt files that are not present in the bundle.

4 / 5

Workflow Clarity

Clear sequence (何时用 → 输入 with an explicit type-inference rule → 步骤 → 产出物) with conditional reads and a signaled completeness gate ("--complete", "EPUB_SEARCH_INCOMPLETE:", "完整性门禁"). Not 5 because pagination/validation specifics and error-recovery steps are deferred to a file absent from the bundle.

4 / 5

Progressive Disclosure

References are one level deep and conditionally signaled ("用户给单图或权要时再 Read prompts/derived_query.md"), but scored against the actual bundle: none of the referenced files (prompts/patent_search.md, prompts/derived_query.md, prompts/covers_rank.md, tools/cnipa_search.py, tools/emit_search_report.py) exist in the bundle, and the single bundle file present (references/patent_type_search.yaml) is never referenced from the body.

3 / 5

Total

16

/

20

Passed

Description

70%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.

The description names a distinctive niche and enumerates concrete searchable fields plus a derived-keyword capability, making it specific and low-conflict. Its main gaps are the total absence of an explicit "use when" trigger clause and missing common synonyms (专利检索、专利查询、公开号), which cap completeness and trigger quality.

Suggestions

Add an explicit trigger clause, e.g. "Use when the user wants to search Chinese patent publications by inventor, applicant, classification, title, or abstract, list a person's publications, or derive search terms from a single figure or a claim set."

Include common natural synonyms users say: 专利检索、专利查询、查专利、公开号、申请号.

DimensionReasoningScore

Specificity

"著录检索：发明人、申请人、分类号、名称、摘要" enumerates five concrete searchable fields and "也可从单图或权利要求生成名称/摘要关键字后查询" adds a second concrete capability. Minor coverage gaps (no 申请号/公开号 search, no mention of outputs) keep it below the comprehensive anchor 5.

4 / 5

Completeness

The "what" is clear (bibliographic field search on the Chinese patent publication site, plus deriving title/abstract keywords from an image or claims), but there is no "Use when…" clause or equivalent explicit trigger guidance — "也可…" describes an additional capability, not a when, which caps completeness at 3.

3 / 5

Trigger Term Quality

Natural field terms users would actually say — 发明人、申请人、分类号、名称、摘要、权利要求、公布公告 — give good keyword coverage. Missing common variations like 专利检索、专利查询、公开号/申请号, so not the comprehensive anchor 5.

4 / 5

Distinctiveness Conflict Risk

"中国专利公布公告著录检索" pins a clear niche (CNIPA publication gazette bibliographic search) with distinct field-level triggers (发明人、分类号、权利要求), so overlap with other skills is minimal.

5 / 5

Total

16

/

20

Passed

Validation

100%

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

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository

handsomestWei/patent-disclosure-skill

Path

skills/patent-search/SKILL.md

Revision

6935a14

Reviewed

Table of Contents

DiscoveryImplementationValidation

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.

Claim skill