German Legal MCP Server
The German Legal MCP Server provides unified access to legal and academic documents for German, Austrian, and EU legal research.
🇩🇪 German Federal & State Legislation (legis:*)
Search state legislation across all 16 Bundesländer and federal law
Retrieve specific laws/norms by ID (e.g.,
bgb/823), browse tables of contents
🏛️ German Court Decisions (rii:*)
Search federal court decisions (BVerfG, BGH, BVerwG, BFH, BAG, BSG, BPatG) and Bavarian state courts
Retrieve full text or summary (Kurztext/Langtext) in Markdown
🇦🇹 Austrian Law (ris:*)
Search and retrieve Austrian federal/state legislation and case law (Judikatur)
Look up individual paragraphs of consolidated laws and browse tables of contents
🇪🇺 EU Court of Justice (icu:*)
Search and retrieve CJEU judgments, opinions, and orders via InfoCuria by case or CELEX number
📜 EU Legislation (eul:*)
Search directives, regulations, decisions, and treaties via EUR-Lex
Retrieve by CELEX number (e.g., GDPR:
32016R0679) with section support
🏛️ German Parliament (dip:*)
Search Bundestagsdrucksachen, legislative processes (Vorgänge), and Plenarprotokoll transcripts
Retrieve full text including Gesetzesbegründungen
📄 Academic Preprints (arxiv:*)
Search arXiv by keywords, author, title, or category; retrieve metadata, abstracts, and full text
📐 Technical Standards (nautos:*)
Search and retrieve DIN/EN/ISO standards with structured TOC and section-level access
✨ Cross-cutting Features
Partial content retrieval via
sectionparameter (Randnummer ranges, headings, line ranges) to avoid context floodingSave to file via
save_pathfor large documentsPandoc-compatible Markdown output across all tools
Provides tools to search and retrieve scientific preprints from arXiv, including support for fetching metadata, abstracts, and full-text document retrieval via Markdown conversion.
Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@German Legal MCP Serversearch for § 823 BGB and explain the liability requirements"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
German Legal MCP Server
German, Austrian & EU legal research — legislation, case law, parliamentary materials, literature and standards
Production status
Version 3.2.1 provides production-ready provider contracts, application components, MCP projections and public/private distributions. Case-law search pages through every source that supports one and reports each source's own hit total. Third-party portals remain external operational dependencies; scheduled live contracts detect availability or response-shape drift. Subscription sources require valid credentials, licences or institutional access.
A Model Context Protocol (MCP) server for German, Austrian and EU legal research, providing unified access to legislation, case law, parliamentary materials, literature, preprints and technical standards.
The provider layer is also available as typed application components. Consumers do not need to run MCP or parse tool output — they can consume normalized federal and Länder case law directly:
import {
CaseLawClient,
} from '@metaneutrons/german-legal-mcp/components/case-law';
const client = new CaseLawClient();
const page = await client.search({
query: 'DSGVO Schadensersatz',
resourceTypes: ['case-law'],
jurisdictions: ['DE', 'DE-NW'],
limit: 25,
});Legislation uses the same application-facing model:
import {
LegislationClient,
} from '@metaneutrons/german-legal-mcp/components/legislation';
const legislation = new LegislationClient();
const laws = await legislation.search({
query: 'Datenschutzgesetz',
resourceTypes: ['legislation'],
jurisdictions: ['DE-NW'],
});Shared provenance, rights, search and document types are exported from
@metaneutrons/german-legal-mcp/contracts. Every provider has a component entry
and a structured data client. The MCP tools use those same clients; MCP output
is only a presentation layer over the application contract.
Multi-domain databases return discriminated unions. For example, RIS exposes
one client for Austrian case law and legislation; narrow each result through
resourceType before using type-specific fields:
import { component as ris } from '@metaneutrons/german-legal-mcp/components/ris';
const client = ris.createDataClient();
const results = await client.search({ query: 'Datenschutz' });
for (const result of results.results) {
if (result.resourceType === 'case-law') console.log(result.fileNumber);
if (result.resourceType === 'legislation') console.log(result.eli);
}Optional portable capabilities cover tables of contents, authentication and operational status. RIS exposes native legislation TOCs. The German legislation client reports a native TOC where the source supplies one and derives it from the document otherwise. RII is case-law-only and does not advertise a TOC capability. Nautos implements TOC and authentication lifecycle capabilities while keeping its session details behind the provider boundary.
Supported Sources
Source | Status | Prefix | Authentication |
Bundes- & Landesrecht | ✅ Available |
| None (public) |
✅ Available |
| None (public) | |
✅ Available |
| None (public OGD API) | |
✅ Available |
| None (public) | |
✅ Available |
| None (public) | |
✅ Available |
| Public key included | |
✅ Available |
| None (public) | |
✅ Available |
| Required (IP or credentials) |
Related MCP server: German Law MCP Server
Features
Bundes- & Landesrecht (legis:* tools)
Federal and state legislation — BUND (all federal laws) + 16 Länder (all states)
No authentication — free public access; the client defines no explicit request limit
Unified interface — one set of tools for all jurisdictions
Full text search — search across state legislation (Länder only)
Resilient input — BUND accepts "§ 823", "823", "Art. 1", "Paragraph 51"
Pandoc-compatible Markdown — clean conversion with Turndown
Save to file —
save_pathparameter to avoid context pollutionAvailable states: BUND, BB, BW, BY, BE, HB, HE, HH, MV, NI, NW, RP, SL, SN, ST, SH, TH
Rechtsprechung im Internet (rii:* tools)
Federal court decisions — BVerfG, BGH, BVerwG, BFH, BAG, BSG, BPatG (from 2010)
Bavarian state courts — AG, LG, OLG, VG, VGH, FG, ArbG, LAG, BayVerfGH via gesetze-bayern.de
NRW state courts — decisions from the official NRWE database via
source: "NW"Lower Saxony state courts — decisions from NI-VORIS via
source: "NI"Brandenburg state courts — decisions from the official Brandenburg decision database via
source: "BB"Bremen state courts — official Bremen VG archive via
source: "HB"; the Bremen index links separate OLG/OVG/VG/LAG portals, so coverage is explicitly partial until those portals expose a common search interfaceSaxony state courts — ESAMOSplus WebForms search for the OLG Dresden archive via
source: "SN"jPortal state courts — Baden-Württemberg, Berlin, Hamburg, Hessen, Mecklenburg-Vorpommern, Rheinland-Pfalz, Saarland, Sachsen-Anhalt, Schleswig-Holstein and Thüringen via their official jPortal portals
Shared DecisionAdapter contract — all new state sources normalize IDs, court, date, file number, ECLI, snippets and Markdown retrieval behind the same
rii:*toolsCross-portal search —
source: "ALL"searches every configured decision portal in parallel, deduplicates overlapping decisions, ranks the consolidated result list and reports unavailable portalsNo authentication — free public access
Full text search — search across all federal court decisions
Kurztext/Langtext — summary or full text via
partparameterRandnummern — formatted as
[Rn. 5]{.rn}(pandoc spans)Save to file —
save_pathparameter to avoid context pollution
RIS Österreich (ris:* tools)
Austrian federal, state & case law — broad Bundesrecht and Landesrecht collection search, plus Judikatur (OGH/OLG/LG via Justiz; VwGH, VfGH, BVwG and others via the
courtfilter)Collection semantics:
ris:searchcan return consolidated norms (BrKons/LrKons) and authentic gazette publications; the returnedapplikationidentifies the result type. Abundeslandfilter restricts Landesrecht to that state's consolidated law (LrKons). For state case law useapplication="judikatur"with the appropriatecourt(e.g.Lvwg), notbundesland.
Normalized application client —
RisDataClient.search()restricts legislation results to consolidated law and supports all 9 Bundesländer through normalized jurisdictionsNo authentication — free public Open Government Data REST API (
data.bka.gv.at/ris/api/v2.6)Latest-first —
sort="date"for the newest decisions; Judikatur Rechtssätze link their full decision text (Entscheidungstext) forris:getNavigate & read statutes —
ris:toc law="ABGB"lists the §§ with headings;ris:get_norm law="ABGB" paragraph="1295"returns a single §Surgical retrieval —
ris:get section=…returns only a Randnummer (Rn 5), an Rn range (Rn 5-9), a line range (lines:1-40), or a heading (Spruch) — all token-preservingPandoc-compatible Markdown — Randnummern as
[Rn. 5]{.rn}spans; document HTML converted with Cheerio + TurndownStructured metadata — Geschäftszahl, Entscheidungsdatum, ECLI, issuing court/organ
Save to file —
save_pathparameter to avoid context pollution⚠️ Austrian law — for German case law use
rii:*, for German legislation uselegis:*
InfoCuria — CJEU (icu:* tools)
EU Court of Justice case law — judgments, opinions, orders from CJEU and General Court
No authentication — free public access via InfoCuria API
Multilingual — documents available in all EU languages (default: DE)
Flexible case lookup — accepts case numbers (C-476/17), CELEX numbers, or internal IDs
Randnummern — formatted as
[Rn. 5]{.rn}Partial content —
sectionparameter for Rn ranges, headings, or line rangesSave to file —
save_pathparameter to avoid context pollution
EUR-Lex (eul:* tools)
EU legislation — directives, regulations, decisions, treaties (TFEU, TEU)
No authentication — free public access via Cellar REST API and SPARQL
Multilingual — documents available in all EU languages (default: DE)
CELEX lookup — retrieve by CELEX number (e.g., "32016R0679" for GDPR)
SPARQL search — search by title keywords, filter by resource type
Partial content —
sectionparameter for articles (Art. 5), headings, or line rangesSave to file —
save_pathparameter to avoid context pollution
DIP Bundestag (dip:* tools)
Parliamentary documents — Bundestagsdrucksachen (Gesetzentwürfe, Beschlussempfehlungen, Anfragen)
Legislative processes — Vorgänge with status tracking and linked documents
Debate transcripts — full text search across Plenarprotokolle (BT and BR)
Full text retrieval — extracted text including Gesetzesbegründungen, with section support
Public API key included — works out of the box (key valid until end of May 2027, override via env var)
Save to file —
save_pathparameter to avoid context pollution
arXiv (arxiv:* tools)
Preprint search — search by keywords, author, title, abstract, or category
Metadata + abstract — default response without full text fetch (token-efficient)
HTML full text — Markdown conversion for papers from ~2024+ (LaTeXML HTML)
PDF fallback — older papers without HTML return abstract + PDF link
No authentication — free public API; the client defines no explicit request limit (upstream usage policies still apply)
Save to file —
save_pathparameter to avoid context pollution
nautos.de (nautos:* tools)
DIN/EN/ISO standards — search and retrieve technical standards from nautos.de
Two-phase document retrieval — outline (metadata + TOC) first, then sections on demand
Automatic authentication — IP-based login (auto-detected), user-based login fallback
Structured TOC — hierarchical table of contents with section IDs for navigation
File cache — 30-day TTL, persistent across restarts (
~/.local/share/german-legal-mcp/cache/nautos/)Save to file —
save_pathparameter to dump full document to disk
Install in Claude Desktop (one-click bundle)
The easiest way to use this server in Claude Desktop is the packaged MCP Bundle (.mcpb) — no Node.js, no npx, no config file.
Download
german-legal-mcp.mcpbfrom the latest release.In Claude Desktop open Settings → Extensions and drag the
.mcpbonto the window (or use Install…).Optionally set the DIP key or nautos credentials in the extension's settings — everything else works out of the box.
The bundle ships the public, no-authentication sources and is cross-platform (macOS and Windows, Apple Silicon and Intel) — Claude Desktop supplies the Node.js runtime, so a single download works everywhere.
Quick Start with npx
npx @metaneutrons/german-legal-mcpor add your MCP client config (e.g., claude_desktop_config.json):
{
"mcpServers": {
"german-legal": {
"command": "npx",
"args": ["-y", "@metaneutrons/german-legal-mcp"]
}
}
}Environment Variables
Provider Enablement
Variable | Default | Description |
| Platform default | Root directory for logs, sessions, caches, metrics, daemon sockets and locks. |
|
| Structured log level. |
|
| Bundes- & Landesrecht |
|
| Rechtsprechung im Internet |
|
| RIS Austria (federal law + case law) |
|
| InfoCuria (CJEU) |
|
| EUR-Lex |
|
| DIP Bundestag (auto-disabled after 2027-06-01 without own key) |
| Public key | Override the bundled public API key |
|
| arXiv preprint search |
| Auto | nautos.de. Auto-enabled with tenant key or credentials, auto-disabled without. |
nautos.de Configuration
Variable | Required | Description |
| For IP-based | Tenant key (e.g., |
| For user-based | nautos.de account username |
| For user-based | nautos.de account password |
| No | Tenant ID (auto-detected from login response) |
Authentication: IP-based login is tried first (requires GLMCP_NAUTOS_TENANT_KEY). If it fails and credentials are set, user-based login is attempted as fallback.
Tools
Bundes- & Landesrecht
Tool | Description |
| Search state legislation. Parameter: |
| Retrieve a specific law/norm. BUND: |
| Compact table of contents for a law — section numbers and headings. Supports |
| List available jurisdictions with implementation status. |
Rechtsprechung im Internet
Tool | Description |
| Search for court decisions. |
| Retrieve full text by doc ID. |
RIS Österreich
Tool | Description |
| Search the broad Austrian RIS Bundesrecht/Landesrecht collections or Judikatur ( |
| Retrieve a RIS document as Markdown by |
| Retrieve a single § of a consolidated law — |
| Table of contents (Inhaltsverzeichnis) of a consolidated law — its §§ with headings — to navigate before |
InfoCuria — CJEU
Tool | Description |
| Search CJEU decisions and opinions. Returns case numbers, ECLI, dates, and document IDs. |
| Retrieve full text by case number (C-476/17) or CELEX number. Supports |
EUR-Lex
Tool | Description |
| Search EU legislation via SPARQL. Filter by type (directive, regulation, decision, treaty). |
| Retrieve EU legislation by CELEX number (e.g., "32016R0679" for GDPR). Supports |
DIP Bundestag
Tool | Description |
| Search Bundestagsdrucksachen by title. Filter by type (Gesetzentwurf, Anfrage, etc.), Wahlperiode, date range. |
| Retrieve full text of a Drucksache by Dokumentnummer (e.g., "19/27426"). Supports |
| Search legislative processes (Vorgänge) with status and linked Drucksachen. |
| Full text search across parliamentary debate transcripts (BT and BR). |
arXiv
Tool | Description |
| Search preprints by keywords, author, title, abstract, or category. Returns metadata + abstract. |
| Retrieve paper by arXiv ID. Default: metadata + abstract. With |
nautos.de
Tool | Description |
| Search DIN/EN/ISO standards by document number. Returns acCode, title, date, type. |
| Retrieve standard by acCode. Returns outline (metadata + TOC) by default; use |
Token-Efficient Document Retrieval
Retrieval behavior is explicit and provider-specific:
Outline-first — Nautos returns metadata and a table of contents before sections or full-file output are requested.
Focused reads — tools that advertise
sectionaccept the selectors documented in their tool description, such as Randnummer, heading, line or article ranges. Selector formats are not assumed across every provider.File output — every tool that advertises
save_pathrequires an absolute path. If that tool also supportssection, it writes the requested section rather than the complete document.
Other retrieval tools return their documented direct response; for example, arXiv returns metadata and the abstract by default. They are not implicitly converted to the outline-first flow.
Markdown Output
Documents are converted to pandoc-compatible Markdown:
Randnummern:
[Rn. 5]{.rn}(bracketed spans)Footnotes:
^[inline footnote text](pandoc inline footnotes)
Development
npm test # Run tests
npm run test:watch # Watch mode
npm run test:coverage # Coverage report
npm run verify # Complete deterministic release gateLive provider contracts
The default suite never uses the network. Opt-in live contracts validate the current upstream response through the same normalized data clients consumed by applications:
npm run test:live:publicThis runs search → normalized reference → document for the public providers and
for every configured German case-law and legislation source. TOC-capable
legislation sources are checked as well. Gesetze im Internet is the documented
exception: it has no search API, so the live contract retrieves bgb/823
directly and validates the BGB TOC separately.
Live output contains only source, document identifier, title, resource type and content length. Full text is asserted in memory and is never written as a test report or CI artifact.
MCP Inspector
npx @modelcontextprotocol/inspector node dist/index.jsCommit Convention
This repo uses Conventional Commits enforced via Husky + commitlint.
Types: feat, fix, docs, style, refactor, perf, test, chore, ci, build, revert
Scopes: legis, rii, ris, icu, eul, dip, nautos, core, deps, config
Architecture
Manifest-driven providers — startup, help, shutdown and public/private distribution use one checked provider manifest
Cheerio + Turndown for HTML → pandoc Markdown conversion
Zod for input validation
Axios for HTTP requests (Legis, RII, RIS, InfoCuria, EUR-Lex, DIP, arXiv, nautos)
Structured JSON errors — all providers return
BaseError.toJSON()withcode,userMessage,recoveryHint; Axios errors auto-wrapped; DNS failures fail fastConversion validation — all HTML→Markdown providers validate output is non-empty; detects upstream layout changes early
Tools namespaced by source (
legis:,rii:,ris:,icu:,eul:,dip:,arxiv:,nautos:)
License
GPL-3.0 - See LICENSE for details.
Maintenance
Resources
Unclaimed servers have limited discoverability.
Looking for Admin?
If you are the server author, to access and configure the admin panel.
Related MCP Servers
- FlicenseAqualityBmaintenanceProvides access to the official German Federal Legal Information Portal (rechtsinformationen.bund.de) enabling AI agents to search German federal laws, court decisions, and legal documentation with authoritative citations from official sources.522
- Alicense-qualityCmaintenanceEnables querying 6,870 German federal statutes, case law, and legislative preparatory works directly from AI assistants and MCP-compatible clients.1404Apache 2.0
- Alicense-qualityFmaintenanceMCP server for accessing and searching German federal law documents from gesetze-im-internet.de. Provides tools to retrieve specific paragraphs, perform full-text searches, and list available laws.8MIT
- Alicense-qualityBmaintenanceProvides read-only access to current German federal law, supporting legislation search, provision retrieval, and citation handling.Apache 2.0
Related MCP Connectors
Verified, citable German & EU law for any LLM. Daily updates from official sources, hosted in DE.
Temporal search and comparison for official Luxembourg and reviewed EU law, with provenance.
Connect AI to millions of laws and court cases with the Lawstronaut MCP.
Latest Blog Posts
- Who's Calling? MCP Hosts Are an Identity Blind Spot (And the Spec Knows It)By Om-Shree-0709 on .mcpAgent IdentityOAuth 2.1
- Your AI Chatbot Just Exposed Your CEO's Salary to an InternBy Om-Shree-0709 on .Agent IdentityMCP SecurityOAuth Delegation
- Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)By Om-Shree-0709 on .Agentic AiPrompt InjectionWebAssembly
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/metaneutrons/german-legal-mcp'
If you have feedback or need assistance with the MCP directory API, please join our Discord server