Open US Law
A free, machine-readable snapshot of United States primary law. Statutory codes, the United States Code, constitutions, state and federal regulations, court rules and agency guidance, attorney general opinions and federal agency decisions, all normalized into a single schema. 5,081,709 sections in one consistent format.
Total sections
Statute sections, state, territorial and federal
Regulation sections in force, state and federal
State attorney general opinions
What is in the dataset
Each record carries a stable act_id, a Bluebook-style citation, its full hierarchy (title, chapter, section), an act_status field (in_force, repealed, reserved, and related states), a source_url where one exists, and cross-references into the United States Code and the Code of Federal Regulations. The schema has 27 columns in total.
This is snapshot v2026.09.1, dated September 30, 2026, 308 files in all. It carries regulations for 40 jurisdictions alongside the current Code of Federal Regulations, 104,540 state attorney general opinions, past editions of the Code of Federal Regulations, proposed federal rules, Federal Register notices, federal session laws, and decisions from the NLRB, FTC, SEC, MSPB, the Board of Immigration Appeals and the Office of Legal Counsel. Every section appears exactly once. New dated snapshots ship on a quarterly cadence, so you can pin an exact version for reproducibility.
A note on provenance
Every record carries the source_url it was published from: a government publisher, or a body the government designates to publish that law. Sections whose source could not be shown are left out rather than published without one.
Access the data
The Hugging Face dataset is canonical and includes an in-browser preview. The scrapers that build it are on GitHub. A Cloudflare R2 mirror serves a combined Parquet tarball and a manifest with a SHA256 for every file.
Licensing
Three layers, each licensed separately and precisely.
Statutory text: public domain
The text of statutes and constitutions is public domain under the government edicts doctrine (Georgia v. Public.Resource.Org, 2020). We assert no rights over it.
Compilation and structuring: CC BY-NC 4.0
The normalized schema, the citations, the hierarchy, and the assembly of the corpus are released under Creative Commons Attribution-NonCommercial 4.0: free for research, teaching, legal aid, journalism and other non-commercial use. Commercial use needs a licence from us, or use the hosted API. Snapshots before v2026.09 were released under CC BY 4.0.
Scrapers: Apache-2.0
The code that collects and structures the data is licensed under Apache-2.0.
If you use the compilation, attribute it as: Open US Law by Vaquill AI, CC BY-NC 4.0.