German Law MCP Server

▶ Try this MCP instantly via Ansvar Gateway

50 free queries/day · no card required · OAuth signup at ansvar.eu/gateway

One endpoint, one OAuth signup, access from any MCP-compatible client.

Connect

Claude Code (one line):

claude mcp add ansvar --transport http https://gateway.ansvar.eu/mcp

Claude Desktop / Cursor — add to claude_desktop_config.json (or mcp.json):

{
  "mcpServers": {
    "ansvar": {
      "type": "url",
      "url": "https://gateway.ansvar.eu/mcp"
    }
  }
}

Claude.ai — Settings → Connectors → Add custom connector → paste https://gateway.ansvar.eu/mcp

First request opens an OAuth flow at ansvar.eu/gateway. After signup, your client is bound to your account; tier (free / premium / team / company) determines fan-out, quota, and which downstream MCPs are reachable.

Self-host this MCP

You can also clone this repo and build the corpus yourself. The schema, fetcher, and tool implementations all live here. What is not in the repo is the pre-built database — TDM and standards-licensing constraints on the upstream sources mean we host the corpus on Ansvar infrastructure rather than redistribute it as a public artifact.

Build your own: run this repo's ingestion script (entry-point varies per repo — typically scripts/ingest.sh, npm run ingest, or make ingest; check the repo root).

The gesetze-im-internet.de alternative for the AI age.

Query 6,870 German federal statutes -- from the BGB and StGB to the GG, BDSG, and more -- directly from Claude, Cursor, or any MCP-compatible client.

If you're building legal tech, compliance tools, or doing German legal research, this is your verified reference database.

Built by Ansvar Systems -- Stockholm, Sweden

Why This Exists

German legal research is scattered across gesetze-im-internet.de, dejure.org, rechtsprechung-im-internet.de, and DIP Bundestag. Whether you're:

• A Rechtsanwalt validating citations in a Schriftsatz or contract
• A compliance officer checking if a statute is still in force
• A legal tech developer building tools on German law
• A researcher tracing legislative history from Drucksache to statute

...you shouldn't need 47 browser tabs and manual PDF cross-referencing. Ask Claude. Get the exact provision. With context.

This MCP server makes German law searchable, cross-referenceable, and AI-readable.

Example Queries

Once connected, just ask naturally:

• "Was sagt § 823 Abs. 1 BGB über Schadensersatz?"
• "Ist das BDSG 2018 noch in Kraft?"
• "Find provisions about Datenschutz in German law"
• "What EU directives does the BDSG implement?"
• "Which German laws implement the GDPR?"
• "Get the preparatory works for the IT-Sicherheitsgesetz"
• "Validate the citation § 433 BGB"
• "Search for Kündigungsschutz in case law"
• "Compare DSGVO implementation across German statutes"

What's Included

Verified data only -- every provision is ingested from official government sources. Zero LLM-generated content.

See It In Action

Why This Works

Verbatim Source Text (No LLM Processing):

• All statute text is ingested from gesetze-im-internet.de official XML exports
• Provisions are returned unchanged from SQLite FTS5 database rows
• Zero LLM summarization or paraphrasing -- the database contains regulation text, not AI interpretations

Smart Context Management:

• Search returns ranked provisions with BM25 scoring (safe for context)
• Three-tier search strategy: exact citation match → FTS5 full-text → LIKE fallback
• Cross-references help navigate without loading everything at once

Technical Architecture:

gesetze-im-internet.de → Parse XML → SQLite → FTS5 snippet() → MCP response
                  ↑                          ↑
           Provision parser           Verbatim database query

Traditional Research vs. This MCP

Traditional: Search gesetze-im-internet.de → Download XML → Ctrl+F → Cross-reference with Drucksache → Check EUR-Lex for EU basis → Repeat

This MCP: "What EU law is the basis for § 1 BDSG?" → Done.

Available Tools (19)

Core Legal Research Tools (8)

Citation Tools (2)

EU Law Integration Tools (5)

Discovery & Metadata Tools (2)

Data Sources & Freshness

All content is sourced from authoritative German legal databases:

• gesetze-im-internet.de -- Federal Ministry of Justice, all consolidated federal statutes
• rechtsprechung-im-internet.de -- Federal court decisions (BVerfG, BGH, BVerwG, BAG, BSG, BFH, BPatG)
• DIP Bundestag -- Legislative preparatory works (Drucksachen, Plenarprotokolle)
• EUR-Lex -- EU cross-reference metadata

Automated Freshness Checks (Daily)

A daily GitHub Actions workflow monitors all data sources:

Security

This project uses multiple layers of automated security scanning:

See SECURITY.md for the full policy and vulnerability reporting.

Important Disclaimers

Legal Advice

THIS TOOL IS NOT LEGAL ADVICE

Statute text is sourced from official gesetze-im-internet.de publications. However:

• This is a research tool, not a substitute for professional legal counsel
• Court case coverage is limited (5,000 decisions) -- do not rely solely on this for case law research
• Verify critical citations against primary sources for court filings
• EU cross-references are extracted from German statute text, not EUR-Lex full text

Before using professionally, read: DISCLAIMER.md | PRIVACY.md

Client Confidentiality

Queries go through the Claude API. For privileged or confidential matters, use on-premise deployment. Lawyers should consider Bundesrechtsanwaltskammer (BRAK) confidentiality obligations when using cloud-based AI tools. See PRIVACY.md for compliance guidance.

Documentation

• Architecture -- Multi-country adapter architecture
• Accuracy Plan -- Roadmap to production accuracy
• Auto Update -- Automated data freshness system
• Security Policy -- Vulnerability reporting and scanning details
• Disclaimer -- Legal disclaimers and professional use notices
• Privacy -- Client confidentiality and data handling

Development

Branching Strategy

This repository uses a dev integration branch. Do not push directly to main.

feature-branch → PR to dev → verify on dev → PR to main → deploy

• main is production-ready. Only receives merges from dev via PR.
• dev is the integration branch. All changes land here first.
• Feature branches are created from dev.

Setup

git clone https://github.com/Ansvar-Systems/German-law-mcp
cd German-law-mcp
npm install
npm run build
npm test

Running Locally

npm run dev                                       # Start MCP server
npx @anthropic/mcp-inspector node dist/index.js   # Test with MCP Inspector

Data Management

npm run ingest                    # Ingest statutes from gesetze-im-internet.de
npm run ingest:cases              # Ingest case law from rechtsprechung-im-internet.de
npm run ingest:prep               # Ingest preparatory works from DIP Bundestag
npm run auto-update               # Run full update cycle
npm run auto-update:dry-run       # Preview what would be updated
npm run drift:detect              # Detect schema/data drift

Performance

• Search Speed: <100ms for most FTS5 queries
• Database Size: ~300 MB (runtime download, Strategy B)
• Ingestion Coverage: 6,870/6,870 statutes (100% TOC coverage)

More Ansvar MCPs

Full fleet at ansvar.eu/gateway.

Contributing

Contributions welcome! See CONTRIBUTING.md for guidelines.

Priority areas:

• Lower court decisions (Landesgerichte, Oberlandesgerichte)
• Historical statute versions and amendment tracking (Fassungsvergleich)
• Expanded case law coverage (currently 5,000 of ~50,000+ published decisions)
• State law (Landesrecht) for major Bundesländer

Roadmap

• Full statute coverage -- 6,870 federal statutes, 91,843 provisions
• Case law -- 5,000 federal court decisions
• Preparatory works -- 89,423 DIP records (WP 19 + 20)
• EU cross-references -- Extracted from statute text
• Citation parsing -- § 823 Abs. 1 BGB and Art. 1 Abs. 1 GG formats
• Free/Professional tier gating -- Honest messaging for tier limitations
• Expanded case law (full archive, ~50,000+ decisions)
• Lower court coverage (Landesgerichte)
• Historical statute versions (Fassungsvergleich)
• State law (Landesrecht)
• English translations for key statutes

Citation

If you use this MCP server in academic research:

@software{german_law_mcp_2025,
  author = {Ansvar Systems AB},
  title = {German Law MCP Server: Production-Grade Legal Research Tool},
  year = {2025},
  url = {https://github.com/Ansvar-Systems/German-law-mcp},
  note = {Comprehensive German legal database with 6,870 statutes, 91,843 provisions, and EU cross-references}
}

License

Apache License 2.0. See LICENSE for details.

Data Licenses

• Statutes & Regulations: German-UrhG-Section-5 -- German statutory public domain. UrhG §5(1) "Amtliche Werke" excludes statutes, ordinances, official decrees and notices, decisions and officially drafted headnotes from copyright protection; §5(2) extends to other official works published in the official interest. Verified verbatim 2026-05-17 -- see docs/audits/2026-05-17-eu-copyright-statutory-works-batch-1b-DE-IE-IT-NL-ES.md. Catalog entry: German-UrhG-Section-5 in infrastructure/attribution-licenses.json.
• Case Law: rechtsprechung-im-internet.de -- same statutory basis (UrhG §5(1) decisions clause)
• Preparatory Works: DIP Bundestag -- UrhG §5(2) official-interest publications
• EU Metadata: EUR-Lex (EU public domain, Decision 2011/833/EU)

About Ansvar Systems

We build AI-accelerated compliance and legal research tools for the European market. This MCP server started as our internal reference tool for German law -- turns out everyone building for the DACH market has the same research frustrations.

So we're open-sourcing it. Navigating 6,870 statutes shouldn't require a law degree.

ansvar.eu -- Stockholm, Sweden