Legal sources with official primary sources

Printed ·

Skip to main content
Skip to the answer
ExploreWorldAIExploreWorldAI

The data engine is updated daily with thousands of new legal data points that strengthen society's decision support

The legal infrastructure, the whole promise on one page

We run a European register of statutes, judgments and regulation that can be fetched directly — by an AI engine, a search engine, a bot or a person. No crawl budget is needed, no data infrastructure has to be built, and nobody has to contact us first. Every row carries provenance, a reading date and a hash, so a decision made on the material can be audited afterwards.

Share this page

What the register holds today

Swedish statutes
52
Guiding judgments
12,713
EU legal acts
32
Articles with their own address
1,798
US rulebooks
67
Decision agents
14,916
Legal layers
9
Jurisdictions
5
Freshness index
76/100
Last reading
2026-09-10

Four readers, one node

The address is the same whoever asks. What differs is the format, never the content.

ReaderGetsHow
PersonThe page, with citation, freshness and the way to the actPlain address
Search engineThe same text in server-rendered HTML with structured dataPlain address
AI engineThe same node as compact JSON: layer, hash, provenance, reading date?format=json
AgentThe same JSON plus receipt, version and dependencies?format=agent

What is free

Readable HTML has always been free. The same now holds for the machine: a daily, bounded dataset without a key and without payment.

  • Every reading page, always, for everyone — a reading page can never return a payment demand.
  • Today's delta: what changed since yesterday, with address, layer, hash and reading date.
  • The node index: the full address list with layer, hash and freshness per node.
  • The citation layer: canonical link, citation string and hash for every node.
  • The freshness stream: the same numbers the pages show, as a file, updated daily.

Terms: use, cite, index and cache freely. Do not resell the dataset as such. Credit ExploreWorld Legal as the source.

https://legal.exploreworldai.com/api/public/v1/open

What costs

Full body text, bulk extraction, reasoning surfaces and agent surfaces are paid per fetch. No subscription, no token fee, no lock-in and no contract to negotiate.

  • Pay per fetch, prepaid against a balance.
  • New agents get free fetches before the balance needs topping up.
  • The price sits below what it costs to crawl the material yourself.
  • No commitment: stop fetching and you stop paying.

The tiers and the price per fetch:

Tier 4 — EU regulation
0.040 EUR
Tier 3 — Case law
0.030 EUR
Tier 2B — US rules
0.020 EUR
Tier 2A — Swedish statutes
0.015 EUR
Tier 1 — Questions
0.005 EUR

How to start

  1. 1Read freely. No registration, no key, no terms to accept.
  2. 2Provision an agent with a single call — you get your own address and your own key.
  3. 3Scale on the machine surface when volume calls for it, with a prepaid balance.

POST https://legal.exploreworldai.com/api/public/v1/agents/provision

Why it can be trusted

  • Provenance per node: every row points at its official source.
  • Reading date and hash on every response, so a copy can be verified against the original.
  • No generated law — we never rephrase a provision.
  • Deterministic answers: the same question returns the same answer, today and in a year.
  • EU operation and EU ownership, at minimal risk class under the AI Act.
  • An open change journal: every movement in the register is dated and readable.

The machine files

A single HEAD request against any address leads to the whole catalogue. Here it is, file by file.

Onward

FAQ

What is free?
All readable HTML, always, for everyone. On top of that the open machine layer: today's delta, the full node index, the citation layer and the freshness stream — no key, no payment, no rate limit.
What does it cost?
Pay per fetch on the machine surface, prepaid against a balance. No subscription, no token fee, no commitment. The price per fetch is published and set below what it costs to crawl the material yourself.
How fresh is the material?
Every layer carries its own reading date and freshness index, and every node carries its own date. The numbers are on the page and in a file, and the change journal shows every movement day by day.
May I use the material commercially?
Yes. The open layer may be used, cited, indexed and cached freely, commercially included — what is not allowed is reselling the dataset as such. Full body text and agent surfaces require an agent and a balance.
How do I start?
Read freely without registering. If you need the machine surface, provision an agent with a single call and get your own address, your own key and free fetches to begin with.

Next step

Three ways to put the register to work in your own practice.

Book a walkthroughPricing and termsAPI and datasets

Start with your task