Awesome Legal Data

Awesome Legal Data is curated by the Open Legal Data project, an initiative that works to make court decisions and legal information freely available, and it gathers in one place the datasets and tools that support legal text processing. Coverage spans several jurisdictions and document types: case law and court records, legislation and statutes, contract collections, and the benchmarks used to test legal language models. Each entry is a description and a link to the primary source, so the list is a signpost, not a data host.

You use it as a map. Open the repository on GitHub, read through the sections, and follow the links for the jurisdiction and document types that match your project. Because it is maintained through pull requests, the commit history and open issues give you a quick read on how current it is, and starring the repo keeps you aware of additions. It pairs naturally with Open Legal Data's own datasets and API, which are worth a look if you need German or European court decisions specifically.

For anyone building a legal RAG system, this is the sensible first stop: it saves you from rediscovering the same handful of foundational sources and helps you see what is available before you commit to an ingestion pipeline. Once you have your bearings, you move to the specific datasets, such as national free-law services or contract corpora, that fit your use case.

The caveats are important in the legal domain specifically. Licensing varies enormously by jurisdiction and dataset, and 'freely readable' often does not mean 'freely reusable', so the list's own note to check each source is one to take seriously. The list carries no blanket licence of its own, and the terms of the material it links to differ from entry to entry, so confirm reuse rights at the source before you index anything commercially. Alongside Awesome LegalTech, which leans towards tools and companies, this list is the more data-focused of the two.