prior-art-search

Search patent databases and academic literature for prior art relevant to an invention. Use when user says "现有技术检索", "prior art search", "专利检索", "check patents", or wants to find relevant prior art.

Quality

—

Does it follow best practices?

Impact

—

No eval scenarios have been run

Securityby

Advisory

Suggest reviewing before use

Quality

Content

80%

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

A concise, highly actionable skill body with a clear workflow, weakened only by missing validation checkpoints for its batch fetching and a dangling reference to a shared-references file that is not bundled.

Suggestions

Add an explicit verification checkpoint after fetching (e.g., confirm each patent/paper's bibliographic data and mark unverifiable items with [VERIFY]) before assembling the report, with a retry-on-failure loop.

Either bundle ../shared-references/prior-art-databases.md or inline the IPC/CPC classification guidance so the dangling reference does not break navigation.

Add a brief per-reference validation step in Step 4 that flags fabricated or low-confidence citations for re-search rather than only relying on a post-hoc Key Rule.

Dimension	Reasoning	Score
Conciseness	Lean and operational: tight constants, concrete search-string templates, and a copy-paste output schema, with no padded explanations of concepts Claude already knows.	3 / 3
Actionability	Provides fully concrete, executable guidance including specific WebSearch/WebFetch query patterns, per-patent extraction fields, and a complete markdown report template ready to fill in.	3 / 3
Workflow Clarity	A clear 6-step sequence exists, but this batch operation (many patents/papers) lacks explicit verification checkpoints and a validate-fix-retry loop, capping it at 2.	2 / 3
Progressive Disclosure	Well-organized into clear one-level sections, but the sole external reference (../shared-references/prior-art-databases.md) points to a non-existent bundle file, a real navigation defect.	2 / 3
	Total	10 / 12 Passed

Description

90%

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, well-targeted description with explicit bilingual triggers and a clear what/when structure; its only weakness is modest action specificity. Voice is correctly third person.

Dimension	Reasoning	Score
Specificity	Names a concrete domain (patent databases and academic literature) and search actions, but frames them as one goal rather than enumerating multiple distinct specific actions.	2 / 3
Completeness	Clearly states what the skill does and includes an explicit 'Use when' clause with concrete triggers, answering both what and when.	3 / 3
Trigger Term Quality	Provides broad natural-language coverage a user would actually say, including bilingual variations (现有技术检索, prior art search, 专利检索, check patents) without missing common forms.	3 / 3
Distinctiveness Conflict Risk	Occupies a clear narrow niche (patent prior-art searching) with distinctive bilingual triggers unlikely to fire for general literature-search skills.	3 / 3
	Total	11 / 12 Passed

Validation

87%

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 — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
allowed_tools_field	'allowed-tools' contains unusual tool name(s)	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	14 / 16 Passed

Repository: wanshuiyin/Auto-claude-code-research-in-sleep
Commit: 82076e5

Reviewed: 5 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.