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

82%

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

75%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.

A well-structured, actionable workflow with concrete search commands and a built-in anti-fabrication checkpoint. Main gaps are the missing validate-retry loop and an external reference that is not part of the bundle.

Suggestions

Add an explicit feedback loop after Step 4/5 (e.g., re-verify [VERIFY]-marked references and re-run searches that returned no results before writing the report).

Inline only the report skeleton and move the full table/example rows into a references/report-template.md file, or confirm the shared-references/prior-art-databases.md file is shipped with the bundle.

Provide at least one fully-resolved example WebSearch/WebFetch query (real keywords) alongside the parameterized templates to lift actionability toward copy-paste-ready.

Dimension	Reasoning	Score
Conciseness	Lean and efficient with constants, a sequenced workflow, and a concrete output template; no padding with concepts Claude already knows, though the inline report template and disclaimer could be marginally trimmed.	4 / 5
Actionability	Provides concrete WebSearch/WebFetch query strings, IPC/CPC guidance, assignee-search tactics, and a full report template; the query examples are parameterized templates with [keywords] placeholders rather than fully copy-paste-ready commands.	4 / 5
Workflow Clarity	A clear six-step sequence with a verification/anti-hallucination checkpoint via the 'Never fabricate … Mark uncertain references with [VERIFY]' rule, so the batch-operation cap is not triggered; lacks an explicit validate-fix-retry feedback loop.	4 / 5
Progressive Disclosure	Well-organized into Constants/Inputs/Workflow/Key Rules with one clearly signaled one-level reference, but no bundle files exist and the referenced shared-references file lies outside the bundle and is absent, leaving the full report template inlined.	4 / 5
	Total	16 / 20 Passed

Description

90%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, third-person description with explicit 'Use when' trigger guidance and bilingual natural keywords. Its only weakness is that it names a single action (search) rather than several distinct capabilities.

Dimension	Reasoning	Score
Specificity	Names the domain (patent databases, academic literature) and the concrete action of searching for prior art, but offers essentially one action applied to two sources rather than several distinct actions, matching the '1-2 concrete actions' anchor.	3 / 5
Completeness	Explicitly answers both what ('Search patent databases and academic literature for prior art relevant to an invention') and when ('Use when user says …'), with concrete trigger phrases.	5 / 5
Trigger Term Quality	Comprehensive natural trigger coverage across two languages — '现有技术检索', 'prior art search', '专利检索', 'check patents', plus 'find relevant prior art' — including synonyms a user would naturally say.	5 / 5
Distinctiveness Conflict Risk	Clear patent/literature prior-art niche with distinct bilingual triggers and minimal overlap risk with other skills.	5 / 5
	Total	18 / 20 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
Path: skills/prior-art-search/SKILL.md
Commit: f4f20f9

Reviewed: 1 day 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.