prior-art-search

Search patent databases and academic literature for prior art.

Updated Apr 29, 2026
One-click install
npx skills add https://github.com/jkfee/Auto-Research --skill prior-art-search-jkfee
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: prior-art-search
Source: https://github.com/jkfee/Auto-Research/tree/main/skills/prior-art-search
Command: npx skills add https://github.com/jkfee/Auto-Research --skill prior-art-search-jkfee

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps teams systematically uncover existing patents and scholarly literature relevant to a new invention, reducing the risk of novelty challenges and enabling informed R&D decisions.

Core Features & Use Cases

  • Systematic patent search: Automates queries across Google Patents, Espacenet, and academic databases to surface relevant patents and papers.
  • Structured classification: Extracts IPC/CPC classifications and key claims; organizes results by relevance and overlap risk.
  • Use Case: A startup developing a new device can quickly assemble a landscape report showing potentially blocking patents and related literature.

Quick Start

Provide a concise invention description to kick off the prior-art search and return a landscape of patents and scholarly papers.

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 description across patent databases and academic literature?

To find prior art, provide a concise invention description to trigger automated cross-database queries across Google Patents, Espacenet, and academic databases. The search extracts titles, abstracts, inventors, and assignees, classifying results by IPC/CPC codes and relevance to uncover potential overlaps.

Can I use this to generate a preliminary freedom-to-operate landscape report?

Yes, you can generate a preliminary freedom-to-operate landscape report by analyzing the structured search results. The report outlines potential blocking patents, non-patent literature overlaps, and classification data, helping teams systematically reduce the risk of novelty challenges for R&D decisions.

How does patent classification by IPC and CPC codes work in these search results?

Patent classification by IPC and CPC codes organizes extracted patent results into structured technological categories. This systematic classification groups related patents and scholarly papers together, allowing you to quickly assess overlap risk and identify relevant prior art within specific technological domains.

What is the best way to search scholarly literature and patents to reduce novelty challenges?

The best way to reduce novelty challenges is executing cross-database queries that simultaneously search patent databases and academic literature. This approach surfaces both patent claims and non-patent literature, providing a comprehensive landscape report of prior art relevant to your invention.

Do I need a detailed technical document to start a prior art search?

No, you do not need a highly detailed technical document to start a prior art search. Providing a concise invention description is sufficient to initiate the automated queries, which then extract and structure the necessary bibliographic data, claims, and classifications for your landscape report.

What are the limitations of automating prior art searches across multiple databases?

Automating prior art searches provides a structured landscape of patents and scholarly literature but remains a preliminary freedom-to-operate consideration. It extracts bibliographic data and classifications to identify potential overlaps, though it does not replace formal legal clearance for definitive IP decisions.