We're building open source tools to unlock the legal knowledge trapped inside documents, contracts, and systems — and making it available to developers, researchers, and anyone who needs it.
Explore the projectsContracts govern nearly every meaningful transaction in modern life. Employment terms, rental agreements, insurance policies, software licenses — these documents determine who owes what to whom, and under what conditions. Yet the tools for understanding, comparing, and analyzing them have remained locked behind commercial paywalls, inaccessible to the people most affected by their contents.
The legal profession has long treated its documents as proprietary artifacts. The irony is hard to miss: the very system meant to codify public rights and obligations operates on infrastructure that most of the public — and most developers — cannot touch.
We believe this has to change. Not through disruption or rhetoric, but through the quiet, compounding power of open source. One tool at a time. One contribution at a time. One knowledge base that anyone can fork, annotate, and extend.
From document parsing to AI-powered annotation, these projects form a complete open source toolkit for legal knowledge work.
Most knowledge lives in documents. Contracts, regulations, research papers, policies — the material that governs how organizations and societies actually work. That knowledge is usually trapped: locked in PDFs, scattered across drives, understood fully by a handful of people who happened to read the right things at the right time.
Then large language models arrived, and the world suddenly needed exactly what careful curation has always produced: structured, annotated, version-controlled knowledge bases that AI can actually reason over. The collaborators these platforms were designed for finally showed up — they just turned out to be AI agents.
But the best AI systems still need carefully curated data. The difference now is that curation and AI can happen in the same place. Human annotation remains the ground truth. AI builds on top of that work — it doesn't replace it.
This is the DRY principle applied to institutional knowledge: annotate once, build on it forever. Fork a public corpus to refine someone else's annotations. Contribute back. Let the community compound what any individual couldn't do alone.