Skip to main content

US laws for your AI agents, your legal AI, your research platform and your compliance tools.

LexBharat is an open API for US primary law: statutes, regulations, court rules and agency guidance, with source links.

Get API key Book a demo

Free credits for personal projects, student work, and academic research.

5.4M+1
primary-law sections
562
US jurisdictions
3
access paths: REST, MCP and SQL
✳Available in an AI assistant →

Trusted by 1,300+ accounts across 150+ organizations

  • Acme Legal
  • Northwind Compliance
  • Lorem & Ipsum LLP
  • Fictional Labs
  • Sample Policy Co.
  • Demo Lending
  • Placeholder Law PLLC
  • Example Firm Ltd.

In their words

“Placeholder testimonial: this is where a customer would describe how verifiable source links changed their workflow.”
Jane Placeholder · Attorney, Example Law PLLC
Read the story

What builders said when the dataset was open-sourced.

  • Sam Sample

    Founder, Demo Labs

    3w

    Finally a clean, citable source for state statutes. Pulling it in this weekend.

    Like · Reply · 42 reactions

  • Riya Example

    Legal engineer

    3w

    The consistent format across 50 states is the part nobody else bothered to build.

    Like · Reply · 18 reactions

  • Alex Placeholder

    Staff engineer, Acme Legal

    2w

    Open-sourcing the dataset is a big deal for anyone building legal AI. Thank you!

    Like · Reply · 27 reactions

  • Jordan Demo

    Attorney

    2w

    Source links on every section make verifying an answer a ten-second job.

    Like · Reply · 9 reactions

  • Maya Fictional

    Product, Sample Policy Co.

    1w

    Tried the MCP server with my agent and the citations came back first try.

    Like · Reply · 31 reactions

  • Chris Lorem

    Researcher

    1w

    Great for academic work: reproducible, versioned, and easy to cite.

    Like · Reply · 14 reactions

  • Dana Mockford

    Compliance lead, Demo Lending

    1w

    We replaced three scrapers with one API call. The change alerts alone are worth it.

    Like · Reply · 22 reactions

  • Priya Sample

    Founder, Fictional Labs

    5d

    Fresh data, real source links, and a format my agents actually understand.

    Like · Reply · 36 reactions

Every US jurisdiction, in one consistent format.

Statutes
Statutes + regulations
Statutes + regulations + constitution

60,494
US Code sections
222,311
CFR sections
196,816
Federal Register rules
2M+
State statute sections
425,427
Statutes at Large
96,966
Agency guidance documents
100,920
Agency adjudications
20,604
State session laws

Federal agency guidance

How agencies say they read their own rules. Guidance is not binding law, and it is stored separately from law that is.

Federal administrative adjudications

Decisions from executive-branch tribunals, served as they were issued. Check these the way you would check any decision.

Federal courts’ own rules

Local rules from many rule sets, each taken from the court’s own site.

CFR
Refreshed daily
US Code
Refreshed weekly
State regulations
Refreshed monthly

Four ways to reach the corpus.

§

REST

The hosted API.

If your product looks up law a section at a time.

◇

MCP

The same corpus inside an agent.

If you want an assistant to look the law up for you, no code.

▤

SQL over the object store

Query the corpus as a database.

If you would rather hold the data yourself and query it.

∴

Vector index

Skip rebuilding retrieval.

If you are building AI search and do not want to index it yourself.

Try the database claim 

-- Illustrative only. The real site shows a DuckDB query over public files.
SELECT citation, section_title, word_count
FROM read_parquet('https://data.lexbharat.example/placeholder.parquet')
LIMIT 5;
Citation in, section out
Bluebook-style and local citation forms resolve to one exact provision.
Point-in-time
Pass an as-of date to read a section as it stood on a past date.
Change alerts
Watch a section or a corpus. Webhooks and email when the text changes.
Formats per section
Styled HTML, plain text, and the publisher’s own PDF or XML where available.

From key to cited law in one call.

request.sh
curl -G https://api.lexbharat.example/v1/statutes/resolve \
  -H "Authorization: Bearer $API_KEY" \
  --data-urlencode "cite=42 U.S.C. § 1983"

Press “Run (mock)” to see the simulated response.

  1. 1. Get your key
  2. → 2. Call the endpoint
  3. → 3. Ship cited law

Two things you get without building them.

API Playground

Run the API before you write a line of code.
(placeholder illustration)
Learn more →

Law Change Alerts

Know when the law you rely on changes.
(placeholder illustration)
Learn more →

Usage-based credits, not seats.

Pro

Where most builds start.

  • ✓20,000 API credits included every month
  • ✓Up to 2 API keys
  • ✓Attribution required
  • ✓Law change alerts by email
  • ✓Email support

Who this is for

  • · Businesses testing a data backend
  • · Lawyers verifying answers
  • · Developers evaluating coverage
Start with Pro now

Business

For shipping data inside your product.

  • ✓200,000 API credits included every month
  • ✓Up to 10 API keys
  • ✓Invite your team
  • ✓Law change alerts by webhook
  • ✓Point-in-time state law
  • ✓Attribution waived
  • ✓Priority support

Who this is for

  • · Legal tech products
  • · Compliance platforms
  • · AI teams grounding agents
Start with Business

Data licensing and bulk access

Custom
  • ✓ Full corpus daily sync into your own database
  • ✓ The source PDFs and text files
  • ✓ Train and fine-tune on all of it
  • ✓ Redistribution and white-label terms
  • ✓ Volume pricing, custom SLA, invoicing

Common questions

Placeholder answer: primary law other than case law, across federal and state jurisdictions. See the coverage section above.

Placeholder answer: sign up, then create a key from a dashboard. (No real signup exists in this recreation.)

Placeholder answer: usage-based credits rather than seats.

Placeholder answer: paid tiers are for commercial use; free credits are for non-commercial work.

Placeholder answer: the pitch is saving the ingestion and cleanup work.

Placeholder answer: see the Visual / JSON toggle in the demo above.

Placeholder answer: yes, per the access paths section (not implemented here).

Placeholder answer: the public site describes an open dataset under a non-commercial license.

Placeholder answer: refresh cadence varies by source: daily, weekly or monthly.

Placeholder answer: HTML, text, and source PDF or XML where available.

Placeholder answer: pulled from official publishers, with a source link on each section.

Placeholder answer: the public site points to open-source ingestion code and open data files.

Working in legal aid or civil justice?

Verified nonprofit legal aid and public interest organizations get full access free (placeholder copy).

Apply for free access

Put real US law behind your product.

Build on the LexBharat APIRead the docs