Zefix MCP

Zefix MCP

Enables AI assistants to search the Swiss Central Business Name Index (Zefix) for companies by name or UID, with optional filters, and retrieve full company details including address, legal form, history, and representatives.

Category
Visit Server

README

Zefix MCP

Zefix Search on MCP Marketplace

⚠️ Unofficial MCP server for zefix.ch – Switzerland Central Business Name Index.
This project is not affiliated with or endorsed by the Federal Department of Justice and Police (FDJP) - Federal Office of Justice (FOJ) who are developing and maintaining zefix.ch.

Features and Use-cases

  • Allows your AI-assistant to search for companies incorporated in Switzerland by their name or UID (CHE-nnn.nnn.nnn) with optional filters by canton of incorporation, legal form, location, previous names, etc. (mimics all filters in Zefix GUI).
  • Equips your AI-assistant with full company details available in Zefix (business names and identifiers, purpose, address, auditors, change history, capital, names of legal representatives, merges & acquisitions) to build further automation workflows or answer business-specific questions.

Sample questions MCP can answer

Company lookup

  • "Find if Google has a legal body in Switzerland and show me their registered address."
  • "Was ist der vollständige rechtliche Name und die UID des Unternehmens, das als Migros bekannt ist?"
  • "Recherche le CHE-169.865.482 et dis-moi tout ce que tu sais à son sujet."

Filtering & discovery

  • "List all active Treuhand companies headquartered in canton Schaffhausen."
  • "Finde Einzelunternehmen im Bereich Malerei, die irgendwo in der Grossregion Zürich eingetragen sind."
  • "Dresse une liste des adresses et des représentants légaux de toutes les banques en Romandie."

History & changes

  • "What is the new name of the company formerly known as 'SwissAir'?"
  • "Zeig mir die vollständige Namens- und Eigentümergeschichte von Credit Suisse."
  • "Y a-t-il des entreprises qui ont repris ou fusionné avec UBS AG ?"

Due diligence & research

  • "Who is listed as auditor for Zurich Insurance Group?"
  • "Ich stehe kurz vor der Unterzeichnung eines Vertrags mit Hans Müller als Vertreter von Lindt & Sprüngli – ist er dazu bevollmächtigt?"
  • "Vérifie que Bollinger est une entreprise active spécialisée dans la plomberie, et dis-moi depuis combien de temps elle existe et quel est son capital enregistré."

Requirements

  • MCP host application (Claude, LLM Studio, VSCode+GHCP/Cline, Cursor, Dive, LibreChat, DeepChat, Chainlit etc.)
  • LLM that supports Tool calling (GPT-4.1+ Claude Sonnet/Opus, Gemini, Llama 3.1+, Qwen3.5, etc.)
  • (optional) Node.js 24+ (for development or local run without npx)

Installation

As a Claude Desktop Extension (easiest)

  1. Go to the Releases page and download the latest zefix.mcpb file.
  2. Double-click it — Claude Desktop opens automatically.
  3. Click Install.

Restart Claude Desktop if you do not see zefix tool in the list of available tools.

Via npx (recommended for all other MCP hosts)

Add to your MCP host config:

{
  "mcpServers": {
    "zefix": {
      "command": "npx",
      "args": ["-y", "zefix-mcp-unofficial"]
    }
  }
}

npx downloads, caches, and runs the package automatically. No local install required.

From source

git clone https://github.com/your-org/zefix-mcp-unofficial.git
cd zefix-mcp-unofficial
npm install
npm run build

Than add this section to config file of your MCP host application (syntax may vary, check documentation for your MCP host app):

{
  "mcpServers": {
    "zefix": {
      "command": "node",
      "args": ["/absolute/path/to/zefix-mcp-unofficial/dist/index.js"]
    }
  }
}

Tool reference

get_companies

Search the Zefix registry. All parameters except name_or_uid are optional.

Parameter Type Description
name_or_uid string Company name or UID (CHE-nnn.nnn.nnn).
language_key string Response language: en, de, fr, it. Default: en.
cantons string[] Filter by canton codes, e.g. ["ZH", "BE"].
locations string[] Filter by legal seat town names, e.g. ["Zurich", "Bern"].
legalForms string[] Filter by legal form, e.g. ["AG", "GmbH"].
exactSearch boolean Search from start of name (default: true). Set false for substring/wildcard (*) search.
phoneticSearch boolean Enable phonetic/fuzzy matching.
includeDeleted boolean Include deregistered companies (default: false).
includeFormerNames boolean Also search former company names (default: false).

Tips:

  • If you know the UID, use it — it will return the most accurate full-detail result.
  • If exactSearch: true returns no results, retry with exactSearch: false and phoneticSearch: true combination.
  • Results are returned in YAML format, with detail level depending on amount of found companies:
    • 1 result → full details including SOGC publication history;
    • 2–10 → detailed summaries, but without full history and list of legal representatives;
    • 11+ → name/UID/status list only.
  • Results are capped at 100 items.
  • The server communicates over stdio (MCP standard transport) — it won't print anything to the terminal when run directly.
  • When MCP server queries Zefix REST API at https://www.zefix.ch/ZefixREST/api/v1/ (public, no authentication required) it submits User-Agent: zefix-mcp-unofficial for transparency. You may want to change uf if using this MCP as part of your larger solution.

Development and Debugging

Build

npm run build
# Compiled output goes to dist/index.js

Inspect

Use the MCP Inspector to interactively call tools and inspect responses:

npx @modelcontextprotocol/inspector node dist/index.js

The first run will prompt you to install @modelcontextprotocol/inspector. After startup, open the URL shown in the terminal (usually http://localhost:5173) to access the inspector UI.

Workflow:

  1. Select the get_companies tool in the left panel.
  2. Fill in parameters (e.g. name_or_uid: "Berg Digital").
  3. Click Run and inspect the raw response.

Publishing a new release

Pushing a v* tag triggers the GitHub Actions release workflow, which runs two jobs in parallel:

Job What it does Where it lands
Publish to npm npm run build, npm publish npmjs.com
Publish Claude plugin npm run build, npx @anthropic-ai/mcpb pack, uploads the .mcpb asset GitHub Releases
# 1. Bump version in BOTH files (npm will reject publishing over an existing version):
#    - package.json  → "version": "x.y.z"
#    - manifest.json → "version": "x.y.z"

# 2. Commit, tag, push — the tag MUST point to the version-bump commit:
git add package.json manifest.json
git commit -m "chore: release vX.Y.Z"
git tag vX.Y.Z
git push && git push --tags

⚠️ Common mistake: creating the tag before editing package.json/manifest.json. CI publishes whatever version string is in package.json at the tagged commit — the tag name itself is ignored by npm. If you tagged too early, move the tag to the correct commit:

git tag -d vX.Y.Z                   # delete local tag
git push origin :refs/tags/vX.Y.Z   # delete remote tag
# edit package.json + manifest.json, then:
git add package.json manifest.json
git commit -m "chore: release vX.Y.Z"
git tag vX.Y.Z
git push && git push --tags

First-time setup (one-off):

  • npm token: create an Access token at npmjs.com → add it as a Repository secret named NPM_TOKEN in GitHub → Settings → Secrets and variables → Actions; token lifetime is 90 days (max) - update regularly.
  • GitHub releases: no extra setup needed — the workflow uses the built-in GITHUB_TOKEN with contents: write permission

License

Recommended Servers

playwright-mcp

playwright-mcp

A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.

Official
Featured
TypeScript
Magic Component Platform (MCP)

Magic Component Platform (MCP)

An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.

Official
Featured
Local
TypeScript
Audiense Insights MCP Server

Audiense Insights MCP Server

Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.

Official
Featured
Local
TypeScript
VeyraX MCP

VeyraX MCP

Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.

Official
Featured
Local
graphlit-mcp-server

graphlit-mcp-server

The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.

Official
Featured
TypeScript
Kagi MCP Server

Kagi MCP Server

An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.

Official
Featured
Python
E2B

E2B

Using MCP to run code via e2b.

Official
Featured
Neon Database

Neon Database

MCP server for interacting with Neon Management API and databases

Official
Featured
Exa Search

Exa Search

A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.

Official
Featured
Qdrant Server

Qdrant Server

This repository is an example of how to create a MCP server for Qdrant, a vector search engine.

Official
Featured