Use cases

Products built on the Moonlit Data API and integration patterns the data layer supports.

Built on the Data API

Two products built on the Moonlit Data API.

Saga

Saga is a legal AI platform serving enterprise law firms across Europe. The Moonlit Data API is its legal knowledge layer for retrieval, citation traversal, and document lookups.

Grounding the platform in the Moonlit Data Layer, 30 European jurisdictions maintained daily, lets Saga deliver a tailored research experience to customers in every jurisdiction they cover.

Moonlit Platform

Moonlit's own platform runs on the same Data API endpoints documented here. Used every day at the Dutch Supreme Court (Hoge Raad), the Public Prosecutor (Openbaar Ministerie), the Dutch Central Bank (De Nederlandsche Bank), Deloitte, and KPMG.

That includes Luna, the streaming legal research assistant, which is itself exposed as a Data API endpoint pair (initialize_chat and ask_question). Luna runs on the Moonlit Data Layer, returns inline citations, and can be embedded directly rather than rebuilt from scratch.

Common integration patterns

Patterns the Data API supports beyond the products listed above. None require custom endpoints.

Legislation and regulatory monitoring

In-house teams and law firms that track new case law, regulator guidance, and legislative changes use the Data API as a daily feed scoped to their fields of law and jurisdictions. The pattern is a scheduled query against keyword_search with a date range, jurisdictions, and document-type filters; results rolled up into a digest, with full text retrieved on demand via retrieve_document.

Filter discovery uses jurisdictions and portals, document types, and fields of law so users can configure exactly what they want to track.

Compliance and regulatory research

GRC and RegTech products use the Data API to ground compliance decisions in real text, regulations, enforcement actions, and supervisory guidance, instead of static checklists. The pattern is semantic_search_reranked on a natural-language business question, scoped to the relevant jurisdictions and fields of law, with retrieve_document pulling full text and the AI-generated summary into the assessment.

Date-bounded queries with sort_type=1 ensure the user is working from current law rather than a superseded provision.

Cross-border legal research

Firms whose clients operate across multiple jurisdictions use one Moonlit integration to cover all of Europe instead of stitching together country-specific APIs. The pattern is a single semantic_search_reranked call against the target jurisdictions, results grouped by jurisdiction for an apples-to-apples comparison.

The sources filter narrows results to specific court levels (supreme courts only, for instance) when the comparison needs to be at the same instance.

Reference graph analytics

Publishers and analytics products use the reference graph to show how a regulation is interpreted across the data layer and to trace how doctrine evolves over time. The graph carries 26M+ relationships; the GDPR alone has 32,000+ incoming references.

The pattern: take a regulation identifier, call articles to break it into article-level units, then call reference_search per article (or per regulation) with sort_type=3 to surface the most-cited interpreting decisions. Add facets=true for distribution by jurisdiction, court, and date.