prior-art-search

Searches patents and scientific literature for prior art relevant to your invention.

Updated May 25, 2026
One-click install
npx skills add https://github.com/duypham2801/ThS_LLM --skill prior-art-search-duypham2801
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: prior-art-search
Source: https://github.com/duypham2801/ThS_LLM/tree/main/.claude/skills/prior-art-search
Command: npx skills add https://github.com/duypham2801/ThS_LLM --skill prior-art-search-duypham2801

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It helps you reduce novelty risk by quickly identifying existing patents and academic publications that may disclose your invention’s key technical elements, so you can refine claims and prioritize what to argue or redesign.

Core Features & Use Cases

  • Patent prior-art discovery: Searches Google Patents and Espacenet using extracted concepts and predicted IPC/CPC classes to surface relevant documents.
  • Academic literature coverage: Finds related prior publications via Google Scholar, arXiv, and Semantic Scholar (when available).
  • Overlap-risk analysis & reporting: Rates each reference’s potential overlap (HIGH/MEDIUM/LOW) and compiles a structured prior art report with freedom-to-operate considerations.

Use case: You have an invention brief for a novel technical approach and want a structured landscape of what existed before your priority date, including both patents and papers.

Quick Start

Run prior-art-search with your invention description to generate patent/PRIOR_ART_REPORT.md containing a categorized prior art landscape and preliminary FTO assessment.

Frequently Asked Questions about prior-art-search

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I find prior art for an invention brief?

To find prior art, this Skill extracts core technical concepts from your invention brief, applies keyword and IPC/CPC classification searches across patent databases and academic literature, and compiles a structured landscape report with overlap risk ratings.

What is the best way to do a freedom to operate pre-check?

A freedom to operate pre-check involves searching existing patents and academic publications for disclosures matching your technical features, then evaluating the overlap risk of each reference to identify potential IP conflicts before your priority date.

Can I use this patent search for IP landscaping?

Yes, you can use this patent search for IP landscaping by analyzing an invention description to surface relevant documents from Google Patents and Espacenet, producing a categorized prior art report with preliminary FTO considerations.

How does academic literature discovery work for prior art searches?

Academic literature discovery for prior art searches works by extracting concepts from your input and querying scholarly indexes like Google Scholar, arXiv, and Semantic Scholar to find related publications that predate your invention.

What limitations exist when searching IPC and CPC classifications?

A key limitation is that this Skill predicts IPC and CPC classes to target searches but does not fabricate citation data, meaning extracted bibliographic and claim signals depend entirely on the actual documents surfaced from Google Patents and Espacenet.