
In simple English
This tool checks if a brand name is already taken in the European Union. It searches across many languages and spots lookalike names by sound and spelling. It can also read official government rejection letters and write draft replies for you. You can run it on your own computer to keep your ideas private, or connect it to an online service.
What it is
An open-source legal technology framework that combines multilingual search with artificial intelligence agents for trademark analysis. It uses a serverless vector database (LanceDB, a system for searching through text and numbers by meaning) to match brand names and classifications.
What it is good for
Lawyers, brand owners, and builders who want to check trademark availability across the European Union without sending sensitive brand concepts to third-party servers.
- Runs locally with Ollama (a tool for running artificial intelligence models on your own computer) to keep brand research private.
- Supports semantic search across all twenty-four official European Union languages.
- Combines structured database filters with vector similarity search for precise brand matching.
- Requires a Python environment and familiarity with command-line tools to install and run.
- Needs significant hardware or an external API key to generate embeddings and run language models.
- Relies on external government portals and data sources that can change their structure over time.
Manual searches on the EUIPO eSearch portal and hand-drafted legal response letters.
Ollama (for local AI inference) and FastAPI (for running the backend service).
- sinonchum/PatentFlow2 stars
A sister project by the same author focused on privacy-first patent analysis and search.
- PrefectHQ/prefect24k stars
Adds scheduling, retries and a run history to Python functions you already have.
- alirezamika/autoscraper8.1k stars
A Python scraper that learns what to extract from one example you give it.
git clone https://github.com/sinonchum/TradeMarkFlow-EU.git cd trademarkflow-eu python3.11 -m venv .venv source .venv/bin/activate pip install -e .