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 materials scientists

Patent and Prior Art Watch Agent

Catch relevant new patent filings every week and explain why each matters.

MonitoringResearchScience and researchProduct development
See how it worksGet access
Patent and Prior Art Watch Agent: what goes in, what the agent does and what you get

What it does

A researcher developing a new process or material usually learns about a competitor's patent late, or never. This agent keeps a plain description of the project and its key features. Each week it searches new published patent applications and grants using classification codes and keywords from that description, reads the claims, and compares them to the project. It scores relevance by how much of the project the claims appear to cover. For borderline cases it pulls the full text and rechecks the claims against the description. It produces a short weekly list with a reason for each score. It gives no legal opinion and never contacts counsel. The scientist approves what goes to counsel. Edge case: a filing with broad claims but unrelated examples is marked for a second look.

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, continueApprovedYes, continueNoNo 1 STARTS WHEN Weekly run 2 USES A TOOL Search new filings by classification and keywords 3 DOES Remove duplicates and filings already reviewed 4 USES A TOOL Read claims and abstracts 5 DOES Compare independent claims to the project features 6 CHECKS THE RESULT Is the score clearly high or low? If not: fetch the full text and recompare claims withthe description. Back to step 4. 7 DOES Write a ranked list with the reason for each score 8 YOU APPROVE Scientist chooses what goes to counsel 9 USES A TOOL Update the watch list and adjust search terms basedon the choices 10 CHECKS THE RESULT Did the search miss a filing that counsel or thescientist already knew? If not: widen the classification codes and keywords,then rerun. Back to step 2. 11 RESULT Weekly watch report
Read the steps as a list
  1. Weekly run
  2. Search new filings by classification and keywords
  3. Remove duplicates and filings already reviewed
  4. Read claims and abstracts
  5. Compare independent claims to the project features
  6. Is the score clearly high or low?If not: fetch the full text and recompare claims with the description. Back to step 4.
  7. Write a ranked list with the reason for each score
  8. Scientist chooses what goes to counselThe agent waits here for your OK.
  9. Update the watch list and adjust search terms based on the choices
  10. Did the search miss a filing that counsel or the scientist already knew?If not: widen the classification codes and keywords, then rerun. Back to step 2.
  11. Weekly watch report

How it decides

It scores a filing high when at least one independent claim covers several key features of the project. Borderline scores trigger a full-text read before the filing is placed on the list.

  • Score high when an independent claim covers 3 or more key features
  • Read full text for any score between 30% and 70%
  • Skip filings already reviewed unless claims changed
  • Widen the search when a known filing is missed

Make it yours

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

  • Classification codes and keywords
  • Score threshold for counsel (default 70%)
  • Weekly run day
  • Number of features that count (default 3)
  • Patent offices to search

What keeps you in control

It always asks you first

  • Anything sent to counsel
  • Changes to the project feature list

Hard limits

  • Never give a legal opinion on infringement
  • Never share the project description outside the team

It stops when

  • Done: weekly list approved
  • Stop: patent search access fails

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 happensThe weekly search found 63 new filings. After removing 12 duplicates, 4 scored above 70%. One claimed a sintering process with the same two temperature steps. Another scored 55%, so the agent read its full text and lowered it to 25% because the claims were limited to polymer binders. The scientist sent the first to counsel.

More agents for materials scientists

Quality control · Science and research

Research Data Intake Validation Agent

A validated, reproducible working dataset with raw data untouched.

7 steps2 checks14 jobs
Research · Document handling

Research Literature Update Agent

An up-to-date evidence map with every new study screened.

7 steps1 check11 jobs
Risk management · Science and research

Lab Inventory Experiment Readiness Agent

Every experiment starts with valid materials and bookings ready.

8 steps2 checks9 jobs
Quality control · Science and research

Scientific Unit Consistency Agent

An analysis working copy in which every calculation uses consistent, documented units.

7 steps2 checks13 jobs