MCP App Store
by KeenEthics
EULEX.AI icon

EULEX.AI MCP Server for ChatGPT

By Eulex.ai

Overview

EULEX.AI supports legal research using EU law from EUR-Lex, Croatian legislation and court decisions, French codified law, and national legislation from Germany, Slovenia and the Netherlands. Search primary sources, retrieve provisions, compare jurisdictions, and inspect available versions, legal status and official publications with source references. Open retrieved articles and Croatian decisions in an interactive source viewer, with exact paragraph or point focus and same-version document context where supported. Guided workflows support contract-clause review, source-based compliance checklists and point-in-time research. The plugin also retrieves reported national measures implementing EU directives and Eurostat statistics. Coverage and available tools vary by jurisdiction; check live coverage before research. EU document text is available in English and national texts generally retain their source language. An EULEX account with an available usage allowance is required. Results support legal research and are not legal advice; verify important conclusions against official sources.

How to connect

  1. ChatGPT: Open the EULEX.AI listing in the ChatGPT app directory and choose Connect. ChatGPT listing EULEX.AI exposes 21 tools here.

Tools

about

ChatGPT
Describe EULEX legal-source coverage, server version and available corpus statistics. Use for questions about what EULEX covers or where its legal information comes from; this does not retrieve the law needed to answer a substantive legal question. One server, N jurisdictions: Croatian legislation + case law (@hr), EU law / EUR-Lex (@eu), French codes / Légifrance (@fr), German federal legislation (@de), Slovenian legislation (@si) and Dutch legislation (@nl). Call list_scopes for per-scope capabilities and identifiers.

browse_taxonomy

ChatGPT
Browse Croatian legal subject areas and their classified regulations, or the catalogue of available French codes. Useful when the relevant Croatian law area or French code is not yet known. For a substantive legal question with known scope, search retrieves candidate provisions more directly. @hr: law-areas tree (481 areas). Without area_id: depth=1 lists top-level areas, depth>=2 lists all areas flat (with parent_law_area_id) in pages of limit — follow has_more and next_offset until the catalogue is complete. With area_id: the area plus its direct children, always; include_regulations adds the regulations in that area (paginated, regulations_has_more / regulations_next_offset). @fr: flat list of all LEGI code titles with article counts (~78 — includes annexes and old/new code variants).

eu_transposition

ChatGPT
Find the national measures a member state reported as implementing an EU directive and its available transposition deadline. Useful for questions such as which Croatian laws implement NIS2, or how a directive was implemented in Germany. Covers reported measures across EU member states; the availability of a notification does not mean that the national law's text is available in EULEX. Reported measures do not prove complete, correct or timely transposition. Shows which national laws each member state notified to the European Commission as implementing (transposing) a directive, and the transposition deadline. Only directives are transposed: EU regulations (CELEX with R, e.g. GDPR 32016R0679) apply directly, so for them the response has applicable false and a note. Data comes from EUR-Lex (Cellar): a first call typically takes 3 to 10 seconds, repeated calls are cached, and a timeout error is retryable once. Reported measures do not establish complete or correct transposition. Response: eu_celex, eu_title, deadline (YYYY-MM-DD) and national_measures, one entry per member state with country_iso2, nim_count, status and its own national_measures list (each with title, notification_date, celex_id, predates_directive, and nn_reference when the title contains a Narodne novine number, usable in find_publication).

eurostat_query

ChatGPT
Retrieve official Eurostat statistics for a requested EU or member-state indicator, period and geography, such as inflation, employment, population or renewable-energy share. Useful when a question needs quantitative evidence, including statistics supporting legal or policy research. It does not establish legal obligations and is unnecessary for a purely statutory question. Official EU statistics, numbers and data tables: GDP, inflation, unemployment, employment, population, migration, trade, energy prices, renewable share, emissions, healthcare, education, digital economy. Use whenever the user needs quantitative data or figures about the EU or its member states. Returns the most relevant dataset as a markdown table plus its source URL — useful for adding quantitative EU data to a legal analysis. The pipeline combines dataset selection with SDMX/Eurostat REST calls and routinely takes 10-25s, so prefer it for questions that genuinely need statistics. The tool is listed in tools/list only on deployments that enable it (the public server does). The response carries available, dataset_code, dataset_title, applied_filters, time_coverage, markdown_table and source_url; when available is false it carries error instead of a table. Check that the dataset title and the applied filters match the requested indicator before quoting a value.

find_publication

ChatGPT
Find Croatian official-gazette publications by NN reference or publication date, or EU documents by entry-into-force year/date. Useful for questions about what was published in Narodne novine during a period, or which indexed EU acts entered into force during a period. EU date filters refer to entry into force, not publication date or the CELEX year. Per-scope modes (provide EXACTLY ONE mode for @hr): @hr: nn_ref='NN 45/2023' (every publication in issue 45 of 2023) or 'NN 45/2023-12' (item 12 of that issue); '2023/45' also works, OR date_from+date_to (ISO dates), OR celex='32016R0679' (EU→HR bridge; currently 0 records until ingest runs). @eu: year=2016 and/or date_from/date_to + optional document_type ('regulation', 'directive', ...) — filters on entry-into-force date. NOTE: entry-into-force year can differ from the CELEX year — acts entering into force retroactively and consolidated versions legitimately appear under a different year than their CELEX id.

get_article

ChatGPT
Read the full wording of a specific legal article, section or supported EU recital, with source references and available version information. Use to quote, explain or verify a named provision, or to check a provision found by search before applying it to a legal question. Accepts supported scoped act names and identifiers, including GDPR, AI Act, Zakon o radu, French codes, GmbHG, ZDR-1 and Dutch acts. Historical article text is available by date for Croatia, France, Slovenia and the Netherlands where indexed; German versions are snapshots, and arbitrary past-date EU article retrieval is unsupported. Returns research text; view_source separately opens an interactive source display.

get_chapter

ChatGPT
Read a complete chapter or part of a Croatian law or bylaw, including its articles. Useful when a question concerns a group of provisions, definitions or exceptions within the same chapter. A single provision fits get_article; a whole act fits get_full_document. Croatia only. Returns one chapter (glava, dio, poglavlje, odjeljak) with the complete wording of its articles. For a single article get_article fits better; for a whole law, get_full_document. The law is identified by URI or by its exact official Croatian title: '@hr/regulation/3244d270-1225-4c41-9567-b684d03c854d' or '@hr/Zakon o radu'. English titles such as 'Labour Act' are not recognised. When several laws share a title, the one in force is used and the older ones are listed in title_resolution. The chapter is given as it is named in the question: a number (3), a Roman numeral (III, IV.a), a label (Glava III, Chapter 3, Dio drugi, Part 2, Odjeljak 1) or words from its title in any grammatical case (Zaštita života, radničko vijeće). A plain number selects the top-level chapter with that number (GLAVA III.); numbered sub-chapters with the same number are listed in chapter_resolution.other_matches and can be fetched by their title words or chapter_ordinal. Response: chapter_label, text (chapter wording with article headings), citation (original Narodne novine number and all amendments, e.g. NN 93/2014, 127/2017, 98/2019, 151/2022, 64/2023), sources per article and chapter_resolution. When no chapter matches, the error lists available_chapters.

get_decision

ChatGPT
Read a complete Croatian court decision by case number, ECLI or a returned decision URI, including its available court, date, reasoning, citations and related decisions. Use to verify or explain a particular judgment and distinguish its facts and holding from a statutory rule. A topic without a known case belongs in case_law_search. Returns research text; view_source separately opens the interactive source display. Accepts the decision's case number (poslovni broj) directly; no search is needed first. Croatian case numbers have the form registry mark, hyphen, number, slash, year, hyphen, suffix: Gž-523/2021-2 (county court, civil appeal), Rev-1356/2024-2 and Revr-1874/2015-2 (Supreme Court revision; Revr in labour cases), P-402/2024-9, Us I-1701/2024-8. Accepted variants: spaces instead of hyphens (Gž 523/2021-2), no diacritics (Gz-523/2021-2), two- digit years (Revr-615/08) and no final suffix (Gž-523/2021, which matches every suffix). ECLI identifiers such as ECLI:HR:VSRH:2009:5567 are accepted too. The court name is not part of the case number. When several courts use the same case number, the response has ambiguous true and candidates (court, date, uri). case_law_search finds decisions on a legal topic. Response: court_name, decision_number, decision_date, ecli, decision_type_name, case_type_name, finality_status, citation (court, case number, date), text (full decision text) and sources.

get_full_document

ChatGPT
Read or summarize the complete indexed consolidated wording of a Croatian, German, Slovenian or Dutch law. Useful when the user asks for a whole act or when provisions throughout an act must be examined; a single-article question fits get_article. Check truncation before claiming complete coverage. Not available for EU acts or French codes through this tool. Country-law responses report truncated=true when the 800,000-character transport limit is reached; use list_articles and get_article for the rest. Use to read or summarize a whole law instead of article-by-article. The response can be long; the in-force version is returned unless a date or version is given.

get_publication

ChatGPT
Open an original Croatian Narodne novine publication or list all items in a specified gazette issue. Use when the user cites an NN reference, asks for the official publication, or needs to connect a published amendment to the affected act. A whole issue contains separate publications, each with its own source. Narodne novine references are written as issue number, slash, year, optionally followed by the item number: NN 14/2019 is issue 14 of 2019 and NN 14/2019-268 is item 268 of that issue. An item reference returns that publication; an issue reference returns every publication in the issue (items, plus a note with their count). Response for one publication: title, publish_date, enter_into_force_date, nn_reference, url (official Narodne novine page) and regulations, the Croatian regulations that this publication created or changed, each with a uri usable in get_regulation, get_article and get_chapter. An amending act or a Constitutional Court decision appears there as a new version of the regulation it changed. Searching by date range is done with find_publication.

get_regulation

ChatGPT
Retrieve the identity and available metadata of a named or retrieved legal act: title, identifiers, source references and recorded status/version information where supported. Useful for confirming that research concerns the correct act and edition. Metadata alone does not establish the wording of a duty, exception or penalty; get_article retrieves the provision.

get_status

ChatGPT
Check a legal act's recorded in-force status and available version or event history. Use for questions about whether a rule is current, repealed or amended, or which recorded Croatian version applied on a date or within a period. Supports EU, Croatia, Germany, Slovenia and the Netherlands with different temporal capabilities. An in-force act does not prove that every provision already applies: application dates, exceptions and transitional rules require reading the relevant provisions. Missing status is unknown; German snapshots are not a complete historical validity timeline. This connection reads EULEX's indexed status and version database only. It does not perform a live request to an external legal-source system.

get_structure

ChatGPT
Locate chapters, sections and article headings within an EU, Croatian, German, Slovenian or Dutch act. Useful for finding where definitions, scope, exceptions or enforcement are located before retrieving their wording. A heading or table of contents is not the full legal rule. @de/@si/@nl return up to 5000 nodes and report has_more if truncated. list_articles supports paginated article discovery. Date selection is available in @si/@nl; @de accepts snapshot version_id only.

list_articles

ChatGPT
List and page through article headings in a Croatian act, French code, German act, Slovenian act or Dutch act. Useful when the article number is unknown or when navigating adjacent provisions. Returns article identifiers rather than their complete wording; use get_article for the text and get_structure for an EU act. @de/@si/@nl: paginated article headings with version-pinned URIs. @si/@nl accept as_of; @de accepts version_id snapshots only. in_force_only is ignored for these scopes; the selected version decides. @hr: all article headings of a regulation version (as_of or version_id supported); the response carries total, has_more and next_offset. @fr: articles of a code ('@fr/code/Code civil') with total, limit and offset — the next page starts at offset + limit while that is below total. @fr rejects as_of and version_id with unsupported_capability.

list_scopes

ChatGPT
Check which jurisdictions and research capabilities are currently available, including historical text, court decisions and source viewing. Use when coverage or support for an operation is uncertain. Configured coverage is EU law, Croatia, French codes, German federal law, Slovenia and the Netherlands; availability does not imply a complete corpus. This is the discriminator tool: call it to learn WHICH scope supports WHICH capability before calling capability-specific tools. A scope missing a capability returns unsupported_capability.

resolve

ChatGPT
Identify a law from its title, common name, abbreviation, CELEX number or official act number, and return its canonical identifier. Use for an unclear or ambiguous act identity or to report the identifier. A known act and article can go directly to get_article; a general legal issue belongs in search. Matching an act does not establish its legal status. Content tools accept names directly ('@hr/Zakon o radu', '@eu/GDPR'), so call them first. Use resolve when the identity is unclear, to report an identifier, or after a content tool returned not_found or ambiguous. '@hr' resolves Croatian titles to regulation UUIDs with ambiguity info. '@eu' first parses the query as a CELEX number ('32023L0970', 'CELEX:32016R0679') or an official act number in English, Croatian, German or French ('Directive (EU) 2023/970', 'Uredba (EU) 2016/679', 'Regulation (EC) No 1907/2006', 'Richtlinie 95/46/EG'), then confirms the document in the corpus and returns its official title, or not_found when no such document exists; a consolidated, corrigendum or national-measure CELEX resolves to its base act with a note. Common-name aliases ('GDPR') follow; any other wording falls back to a semantic title search whose candidates carry similarity, with ambiguous true when none is clear. A match is an identifier, not a legal-status check: confirm in_force with get_status or get_regulation. Returns matches[] with uri, plus ambiguous and best_guess.

view_source

ChatGPT
Display a retrieved legal provision or Croatian court decision in an interactive source viewer when the user asks to open, inspect or read the source or its surrounding context. Shows original wording, source identity, version and an exact paragraph or point where matched. Use a returned source ID and preserve its article and version. This is a presentation tool; search and get_article perform ordinary legal research without displaying a card for every retrieval. Document context is available for EU, Croatia, Germany, Slovenia and the Netherlands; France remains article-only. No supplied or generated text is treated as a retrieved source.

view_source_context

ChatGPT
Load same-version surrounding source text for the interactive source viewer after its context control requests it. App-only retrieval operation, not a general model-facing legal-research entry point. Supports available EU, Croatian, German, Slovenian and Dutch document context and complete Croatian decisions. An app-only retrieval operation used by the interactive viewer. Returns the same source identity and version as the cited article with the surrounding original text where available (EU, HR, DE, SI, NL) or the complete Croatian decision.

Capabilities

Interactive

Example Prompts

Click any prompt to copy it.

Similar Apps

Listed on MCP App Store

Listed on MCP App Store

Add this badge to your site or README. It links to this page.

HTML

<a href="https://mcp-app-store.keenethics-labs.com/app/eulex-ai"><img src="https://mcp-app-store.keenethics-labs.com/badges/listed-on-mcp-app-store.svg" alt="Listed on MCP App Store" width="220" height="44"></a>

Markdown

[![Listed on MCP App Store](https://mcp-app-store.keenethics-labs.com/badges/listed-on-mcp-app-store.svg)](https://mcp-app-store.keenethics-labs.com/app/eulex-ai)