The contract you can't paste into ChatGPT: a red-flag finder that runs in your browser
This is a submission for the Hacktoberfest Open-Source AI Challenge Week 2: Side Hustle What I Built Every freelancer has signed a contract that the client's lawyer wrote. Twelve pages, and somewhere in clause 6.1 you have agreed not to work for anyone in "the beverage industry" for eighteen months, in 4.1 your logo sketches became their property the moment you drew them, and in 2.2 and 2.3 you…
This is a free tool that scans your contract before you sign it, using your own device and web browser. Called Fine Print, it identifies problematic clauses that freelancers often overlook in their contracts with clients. The software highlights issues such as ownership of your work, non-compete clauses, unlimited revisions, and clauses that limit your ability to challenge the client.
It then generates a polite email to the client, with sections and blanks for you to negotiate new terms. Fine Print does not require an account or sign-up, and it operates completely offline, scanning your contract without uploading any data or information. This tool is particularly useful for freelancers who sign contracts occasionally and cannot afford a lawyer for each one, such as designers, developers, writers, and consultants.
It uses EmbeddingGemma, Google's open-weight embedding model, to analyze each clause of your contract in real-time, without any server-side processing. The software is built using static files and does not rely on any backend or API keys, making it a lightweight and easy-to-use tool. Debugging and updates are straightforward, as there is no server-side component to maintain.
Fine Print was developed by Vanshaj Poonia, who wrote the advice for the clauses based on his own knowledge, rather than having the AI generate it. This ensures that the information provided is accurate and reliable. The tool is available for free at https://vanshajpoonia.github.io/fine-print/.
Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.