I've built Patent Connector, a Model Context Protocol server now in open beta. It connects ChatGPT Desktop, Claude Desktop, and other MCP-compatible AI tools directly to patent databases - starting with the free EPO OPS API. The Problem with Web Search Tools like ChatGPT can access Google Patents
I've released a complete Go client for the European Patent Office's Open Patent Services API v3.2. The library provides access to patent bibliographic data, full-text documents, search capabilities, family information, and legal status data. Building on Official Specifications I started with EPO's official
I've released a new Go client library for the European Patent Office Bulk Data Distribution Service. The library provides straightforward access to EPO's bulk patent data products through a clean, production-ready Technical Approach Since no official API specification existed for BDDS, I created an OpenAPI v3
Great news for patent data enthusiasts! The USPTO has launched its Open Data Portal (ODP), making it faster and easier to access patent and trademark datasets. Whether you're a researcher, developer, or just a data geek, this platform simplifies the way you retrieve and customize USPTO data. The
It's been a while since my last post - lots of things are happening in parallel. Besides my day-to-day work, I'm currently building an international patent online community - a place where professionals and anyone interested in patents can meet, share ideas, and chat. On top
Exciting news for patent data users! As of January 1, 2025, the European Patent Office (EPO) has introduced a new pricing policy, making several previously paid bulk data products completely free. These datasets will now be available in the Public Area of the Bulk Data Download Service (BDDS), eliminating the
The European Patent Office has unveiled a powerful new tool for navigating the EPC 2023, PCT-EPO Guidelines, EPC Guidelines, PCT Guidelines, and more. This Legal Interactive Platform comes in the form of an interactive legal chatbot, leveraging large language model (LLM) technology. Originally launched as an internal tool, it’s
The European Patent Office has introduced its new Technology Intelligence Platform, a ready-to-use solution based on Jupyter - a web-based development environment that integrates Python, Markdown, and Git. This platform aims to make it easier to process, analyze, and visualize data, ultimately turning raw information into valuable patent knowledge. Why
The patent sector offers numerous software tools, but as a proponent of open-source and self-hosted solutions, I chose phpIP as the first tool to review. Here’s an overview. phpIP is a web-based platform. Its creators describe it as a web tool for managing an IP rights portfolio, especially patents.
Digital search works so well that it often feels magical—how else could systems sift through vast amounts of data in milliseconds and even seem to “understand” what you’re looking for? But there’s no magic involved—just some preparation behind the scenes. How Search Actually Works When you
Working with patent data often begins with a familiar headache: how to efficiently collect it. With so many data sources offering data in myriad formats—and these formats constantly evolving—it’s a challenge to maintain a reliable system. For instance, the USPTO is transitioning to its new Open Data
Patent numbers may seem straightforward, but they hold far more complexity than many people realize. Each number is like a fingerprint, with layers of information encoded in a unique combination of codes and identifiers. From country codes that change over time to kind codes hinting at a document’s status,
first steps
If you’re thinking about creating a custom patent database, here are the core questions to consider: * Data Sources: Will you use your own patent data, data from patent offices, or datasets from other companies? * Objectives: Define the specific issues your database should solve—special calculations, complex data management, or
gotcha
When someone applies for a patent, they embark on a long journey through filing, examination, grant, and maintenance. This process creates a massive trail of data, inputted at each step by people in multiple systems. This data is processed, transformed, and passed through various networks, databases, and software systems, leading
first steps
Since patents are national rights, the same invention is often covered by multiple patents in different countries. These related patents form a ‘patent family.’ Different data providers use their own methods and criteria to group patents into families, which can lead to variations in how families are defined. Patent families
first steps
Some people think that if someone holds a patent, the invention is protected worldwide. But here’s a key fact to start with: patents don’t automatically apply globally. They’re national rights, meaning if you patent something in one country, it’s not protected in others. To simplify international
Hello, and welcome to my blog! I’m a software developer and consultant with extensive experience in patent data. Here, we’ll explore how patent data, paired with innovative software solutions, can drive new insights and fuel strategic thinking. Let’s hope this won’t be a 4′33″ thing—