MCP App Store
by KeenEthics
Juridata icon

Juridata MCP Server for Claude and ChatGPT

By JURIDATA

Overview

Juridata brings the expertise of Moroccan law into Claude. It searches case law from the Cour de cassation, legislation published in the Official Gazette, and Ministry of Justice resources — and every answer is backed by references you can verify against the source. Built for legal professionals in Morocco — lawyers, in-house counsel, and notaries — Juridata lets you research, analyze, and draft without leaving your conversation. What you can do: • Search case law and legislation, with cited references. • Get reasoned answers to questions of Moroccan law. A pioneering African legaltech: Juridata starts with Moroccan law and is built to scale across the continent — bringing verifiable, jurisdiction-specific legal AI to a market that global tools don't cover. Before you start: an active Juridata account is required.

How to connect

  1. Claude: Open the Juridata listing in the Claude connectors directory and choose Connect. Claude listing Juridata exposes 4 tools here.
  2. ChatGPT: Open the Juridata listing in the ChatGPT app directory and choose Connect. ChatGPT listing Juridata exposes 4 tools here.

Tools

fetch_document

ChatGPT
Fetch the full text of a decision or law by the `id from a search hit. Long documents are returned one window at a time (up to ~20k tokens). When a result has "truncated": true, call this tool again with the SAME document_id and offset set to the returned next_offset to read the next part — repeat until truncated is false. Each result reports offset, returned_chars, total_chars and next_offset so you know how much remains. Args: document_id: An id returned by a search tool, e.g. "juris:1234" for a court decision or "law:567" for a legal text. offset: Character position to start reading from. Use 0 (default) for the start, then the previous result's next_offset` to continue.

fetch_document

ChatGPT
Fetch the full text of a decision or law by the `id from a search hit. Long documents are returned one window at a time (up to ~20k tokens). When a result has "truncated": true, call this tool again with the SAME document_id and offset set to the returned next_offset to read the next part — repeat until truncated is false. Each result reports offset, returned_chars, total_chars and next_offset so you know how much remains. Args: document_id: An id returned by a search tool, e.g. "juris:1234" for a court decision or "law:567" for a legal text. offset: Character position to start reading from. Use 0 (default) for the start, then the previous result's next_offset to continue. Check a law's status and text_asof_date before relying on it as current law (only in_force` is verified); the corpus does not flag overturned decisions.

search_jurisprudence

ChatGPT
Semantic search over Moroccan court decisions (jurisprudence). Args: keywords: An array of one or more search terms or phrases, in Arabic or French (e.g. ["التهرب الضريبي"]). A single string is also accepted. max_results: Maximum number of decisions to return (1–25). court_room: Optional chamber filter (civile, commerciale, sociale, penale, administrative, fonciere, familiale). Omit to search all. year: Optional — restrict to decisions issued in this year, e.g. 2021. Returns ranked hits, each with an `id (juris:<n>), a confidence in [0, 1] (semantic relevance of the excerpt to the query; ~0.7+ is strong, ~0.4 is weak), the decision and file numbers, the court room(s), the date, and a short excerpt. Pass a hit's id to fetch_document` for the full decision text.

search_jurisprudence

ChatGPT
Semantic search over Moroccan court decisions (jurisprudence). Args: keywords: An array of one or more search terms or phrases, in Arabic or French (e.g. ["التهرب الضريبي"]). A single string is also accepted. max_results: Maximum number of decisions to return (1–25). court_room: Optional chamber filter (civile, commerciale, sociale, penale, administrative, fonciere, familiale). Omit to search all. year: Optional — restrict to decisions issued in this year, e.g. 2021. Returns ranked hits, each with an `id (juris:<n>), a confidence in [0, 1] (semantic relevance of the excerpt to the query; ~0.7+ is strong, ~0.4 is weak), the decision and file numbers, the court room(s), the date, and a short excerpt. Pass a hit's id to fetch_document for the full decision text. Hits are similarity-ranked, so they skew toward decisions that agree with each other; to avoid missing a contrary or minority line, run a second search for the OPPOSING holding. The corpus does not record whether a decision was later overturned or superseded — read the date` and do not present a ruling as settled law without that caveat.

search_laws

ChatGPT
Semantic search over Moroccan legal texts (codes, statutes, regulations). Args: keywords: An array of one or more search terms or phrases, in Arabic or French (e.g. ["الضريبة على الدخل"]). A single string is also accepted. max_results: Maximum number of law passages to return (1–25). category: Optional law-category filter. Call list_legal_categories for the valid values. Omit to search every category. Returns ranked hits, each with an `id (law:<n>), a confidence in [0, 1] (semantic relevance of the excerpt to the query; a directly-named law number resolves to 1.0), the law name and category, and a short excerpt. Pass a hit's id to fetch_document`.

search_laws

ChatGPT
Semantic search over Moroccan legal texts (codes, statutes, regulations). Args: keywords: An array of one or more search terms or phrases, in Arabic or French (e.g. ["الضريبة على الدخل"]). A single string is also accepted. max_results: Maximum number of law passages to return (1–25). Returns ranked hits, each with an `id (law:<n>), a confidence in [0, 1] (semantic relevance of the excerpt to the query; a directly-named law number resolves to 1.0), the law name and category, and a short excerpt. Pass a hit's id to fetch_document. There is no category filter: every call searches the whole legislative corpus, so the keywords are the only thing that narrows it, and an empty result cannot be a filter artefact. The category on each hit is descriptive metadata only — the labels are assigned per source document and are frequently wrong, which is why they are not offered as a filter. The on-point provision is rarely the whole rule: search separately for the EXCEPTIONS and derogations that narrow it, not just the article that matches the query. Each hit carries text_asof_date and a status: only in_force (reviewed) supports calling a text current law; repealed / older_edition name their successor; repeal_pending / superseded_pending still applied on the reference date but are replaced from effective_on; unverified` (most texts) means currency is unknown — say so.

fetch_document

Claude

search_jurisprudence

Claude

search_laws

Claude

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/plugin_asdk_app_6a3ba5ef1c448191a4c89868b2688b7e"><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/plugin_asdk_app_6a3ba5ef1c448191a4c89868b2688b7e)