Complete AI Training
Sign inGet my AI kit

Your job's AI kit

Get your AI kit

Tell us who you are and what you do. We show you your kit right away and email you the link: skills, prompts, AI agents, MCP servers and courses for your job.

500+ jobs ready, and we make a kit for any other job. No payment needed to look.

Share

AI agent for patent agents

Prior Art Search Agent

A search report with the closest references charted against features

Prior Art Search Agent: what goes in, what the agent does and what you get

What it does

A weak prior art search leads to wasted filing costs or patents that fail later. For a new invention, this agent builds and runs the search. It reads the invention disclosure, extracts key features, and builds search strings and classification codes. It searches patent and non-patent literature, ranks results by how many features each reference discloses, and builds a feature chart. If the top results miss key features, it refines the terms or codes and searches again. It stops when two rounds add nothing closer. References with all features are flagged. It drafts a search report. The patent agent reviews the results and gives the opinion. Edge case: a reference is in another language, so the agent uses the English family member or a machine translation and notes it.

How it works

Follow the arrows from top to bottom. The orange dashed arrow is the loop: when a check fails, the agent goes back and tries again.

Start and resultWhat it doesA check on its own workWaits for your OKGoes back and retries
Yes, continueApprovedNo 1 STARTS WHEN Disclosure received 2 USES A TOOL Extract features 3 DOES Build search strings and codes 4 USES A TOOL Search patents and literature 5 DOES Chart features 6 CHECKS THE RESULT Do new searches stop finding closer references? If not: refine terms and codes and search again. Back tostep 3. 7 YOU APPROVE Patent agent reviews 8 RESULT Search report
Read the steps as a list
  1. Disclosure received
  2. Extract features
  3. Build search strings and codes
  4. Search patents and literature
  5. Chart features
  6. Do new searches stop finding closer references?If not: refine terms and codes and search again. Back to step 3.
  7. Patent agent reviewsThe agent waits here for your OK.
  8. Search report

How it decides

It ranks references by features disclosed and refines until results stabilize.

  • Reference with all features: flag
  • Foreign language: use family or translate
  • Stop when two rounds add nothing new

Make it yours

Every agent is a starting point. You choose these settings for your own situation.

  • Databases
  • Rounds
  • Chart format
  • Report template

What keeps you in control

It always asks you first

  • Patentability opinion

Hard limits

  • Never gives the opinion
  • Never discloses the invention outside secure tools

It stops when

  • Done: report ready
  • Stop: disclosure unclear

Set it up

We guide you through the set-up, step by step

Members get the full set-up guide for this agent. No technical skills needed: you copy, paste and upload.

10 minto set it up in your AI
5 AIsChatGPT, Claude, Copilot, Gemini, Grok
  • One set of instructions to paste into your AI, with the clicks for ChatGPT, Claude, Microsoft 365 Copilot, Gemini and Grok
  • The agent then walks you through connecting your own data, one source at a time
  • A downloadable copy with the flow chart, the rules and the full guide
Get access to this agent

An example run

What happensA battery cooling disclosure had 6 key features. Round one found references with up to 4. The agent added a new classification code and found a Japanese patent with 5, using its English family member. Round three found nothing closer, and round four confirmed it, so the search stopped. The patent agent reviewed the chart and approved the report.

More agents for patent agents