Swiss Environment MCP Server

🇨🇭 Part of the Swiss Public Data MCP Portfolio

🌿 swiss-environment-mcp

MCP server connecting AI models to Swiss environmental data from BAFU – air quality, hydrology, natural hazards, wildfire danger and open environmental datasets.

🇩🇪 Deutsche Version

---

Overview

swiss-environment-mcp gives AI assistants like Claude direct access to real-time environmental data from Swiss federal authorities – no API keys required. Air quality readings from the national NABEL monitoring network, hydrological gauging stations, natural hazard bulletins, and the full BAFU dataset catalogue are all accessible through a single standardised MCP interface.

The server covers four thematic clusters: air quality (NABEL), hydrology, natural hazards, and the BAFU open data catalogue. Each cluster maps to a group of purpose-built tools that translate raw agency data into clean JSON responses.

Anchor demo query: "What is the current air quality at the NABEL station Zürich-Kaserne – and does it comply with WHO 2021 guidelines?" → More use cases by audience →

---

Features

• 🌬️ Air quality monitoring – 16 NABEL stations, NO₂/O₃/PM10/PM2.5/SO₂/CO, Swiss LRV + WHO 2021 limit checks
• 💧 Hydrology – water levels, flow rates, temperatures across Swiss gauging stations
• 🚨 Flood warnings – active alerts filtered by danger level and canton
• 🏔️ Natural hazard bulletin – SLF/BAFU bulletin in DE/FR/IT/EN, region-specific warnings
• 🔥 Wildfire danger – canton- and region-level fire danger index
• ❄️ Snow & avalanches (SLF) – snow depth, new snow per IMIS station; avalanche danger levels (EAWS)
• 🦌 Hunting statistics – cull & game-loss figures per species, canton and year (federal hunting statistics)
• 📦 BAFU open data catalogue – search and retrieve environmental datasets via CKAN
• 🔑 No authentication required – all data sources are publicly accessible
• ☁️ Dual transport – stdio for Claude Desktop, Streamable HTTP/SSE for cloud deployment

---

Prerequisites

• Python 3.11+
• No API keys needed – all endpoints are publicly accessible without authentication

---

Installation

# Clone the repository
git clone https://github.com/malkreide/swiss-environment-mcp.git
cd swiss-environment-mcp

# Install
pip install -e .

Or with uvx (no permanent installation):

uvx swiss-environment-mcp

Or via pip:

pip install swiss-environment-mcp

---

Quickstart

# Start the server (stdio mode for Claude Desktop)
swiss-environment-mcp

Try it immediately in Claude Desktop:

"What is the current air quality at NABEL station Zürich-Kaserne?" "Are there any active flood warnings in Switzerland right now?" "What is the wildfire danger level in Canton Valais?"

---

Configuration

Claude Desktop

Minimal (recommended):

{
  "mcpServers": {
    "swiss-environment": {
      "command": "uvx",
      "args": ["swiss-environment-mcp"],
      "env": {}
    }
  }
}

Config file locations:

• macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
• Windows: %APPDATA%\Claude\claude_desktop_config.json

After saving, restart Claude Desktop completely.

Cloud Deployment (SSE for browser access)

For use via claude.ai in the browser (e.g. on managed workstations without local software):

Render.com (recommended):

1. Push/fork the repository to GitHub
2. On render.com: New Web Service → connect GitHub repo
3. Render detects render.yaml automatically
4. In claude.ai under Settings → MCP Servers, add: https://your-app.onrender.com/sse

Docker:

docker build -t swiss-environment-mcp .
docker run -p 8000:8000 swiss-environment-mcp

💡 "stdio for the developer laptop, SSE for the browser."

---

Available Tools

🌬️ Air Quality / NABEL (3 tools)

💧 Hydrology (4 tools)

🏔️ Natural Hazards (3 tools)

❄️ Snow & Avalanches / SLF (3 tools)

🦌 Hunting & Wildlife (2 tools)

📊 Environmental Data Catalogue (2 tools)

Example Use Cases

---

🛡️ Safety & Limits

---

Architecture

┌─────────────────┐     ┌───────────────────────────┐     ┌──────────────────────────┐
│   Claude / AI   │────▶│   Swiss Environment MCP   │────▶│  BAFU / Swiss Agencies   │
│   (MCP Host)    │◀────│   (MCP Server)            │◀────│                          │
└─────────────────┘     │                           │     │  hydrodaten.admin.ch     │
                        │  12 Tools · 3 Resources   │     │  naturgefahren.ch        │
                        │  Stdio | SSE              │     │  waldbrandgefahr.ch      │
                        │                           │     │  opendata.swiss (CKAN)   │
                        │  api_client.py            │     └──────────────────────────┘
                        │  server.py (FastMCP)      │
                        └───────────────────────────┘

Data Sources

All data: publicly accessible, no authentication required.
Attribution required: BAFU / SLF (WSL) must be cited as the source when using their data.

---

Project Structure

swiss-environment-mcp/
├── src/swiss_environment_mcp/
│   ├── __init__.py          # Package
│   ├── server.py            # FastMCP server: 12 tools, 3 resources
│   ├── api_client.py        # HTTP client + egress allow-list (SSRF guard)
│   └── logging_setup.py     # structlog -> stderr
├── tests/
│   ├── test_unit.py         # Mocked unit tests (no network) — CI default
│   ├── test_integration.py  # Live API tests (marker: live)
│   └── test_20_scenarios.py # Live scenario coverage
├── scripts/tool_snapshot.py # Tool-definition hash snapshot (rug-pull guard)
├── docs/                    # security.md, scaling.md, roadmap.md
├── .github/
│   ├── dependabot.yml       # Monthly dependency/action updates
│   └── workflows/           # ci.yml, security.yml (gitleaks), live-tests.yml, publish.yml
├── Dockerfile               # Multi-stage, non-root container
├── render.yaml / Procfile   # Cloud deployment
├── tool-snapshot.json       # Committed tool-definition snapshot
├── .env.example             # Non-secret config template
└── pyproject.toml           # Build configuration (hatchling)

Single-module layout (rationale, audit ARCH-011): the 12 tools live in one server.py rather than a tools/ package. They are thin, uniform wrappers over api_client.py sharing the same input/response patterns, so a single well-sectioned module stays more navigable than 4 near-identical files. This is a deliberate, documented deviation; a split is revisited if tool logic grows non-uniform.

---

MCP Protocol Version & Maintenance

• MCP protocol: negotiated at initialize time and handled by the pinned MCP SDK (mcp[cli]). The SDK version is the canonical pin; SDK/protocol updates land via Dependabot PRs (.github/dependabot.yml, monthly).
• Tool-definition stability (audit SEC-022): any change to a tool's name, description or parameters changes tool-snapshot.json; CI fails until the snapshot is regenerated and a CHANGELOG entry + version bump are added.
• Update policy: review Dependabot PRs monthly; bump the version (semver) on any tool-definition or behaviour change.

Lifecycle Phase

This server is in Phase 1 (read-only) — all tools read-only, no auth, no side effects. The phase model and prerequisites for Phase 2 (write/auth) are in docs/roadmap.md. Security architecture (SSRF/egress, secret management, lethal-trifecta assessment): docs/security.md. Scaling/session strategy: docs/scaling.md.

---

Known Limitations

• Hydrology via LINDAS: env_hydro_current, env_hydro_stations and env_flood_warnings query the BAFU LINDAS SPARQL endpoint (typed live values: level, discharge, water temperature, danger level). LINDAS holds current values only (one observation per station) — it is not a historical time series. See docs/probe-lindas-hydro.md.
• Historical hydrology / env_hydro_history (BUG-01 resolved): the old hydrodaten.admin.ch/lhg/az/* REST endpoints (hourly CSV, warnings.json, station JSON) are decommissioned (404). env_flood_warnings now uses LINDAS dangerLevel instead. Real historical time series (daily / long-term means — e.g. summer 2024 vs. long-term average) are not freely available via API; they must be ordered from the BAFU Hydrological Enquiry Service (abfragezentrale@bafu.admin.ch). env_hydro_history returns the latest LINDAS value plus this access path.
• Flood warnings: env_flood_warnings reads LINDAS dangerLevel; a canton filter is not available there (LINDAS carries no canton code), so it is reported but not applied.
• NABEL: Near-real-time data only; no historical time series via this server.
• Natural hazards: Bulletin availability depends on SLF/BAFU publication schedule.
• Wildfire danger: Regional granularity varies by season and data availability.
• Hunting statistics (env_hunting_stats): The jagdstatistik.ch backend is undocumented (a content-negotiated web-app endpoint). A schema-guard degrades gracefully if the structure changes. Species/canton/datatype lookups are embedded (harvested 2026-07-19); figures are fetched live for 2015–2024. Licence (researched 2026-07-19): the data is owned by BAFU (compiled from cantonal offices; site tech by Wildtier Schweiz) and is not published as a licensed dataset on opendata.swiss; no explicit licence is stated on the source. Responses therefore require source attribution to BAFU; formal licence confirmation from BAFU is still pending. See docs/probe-jagdstatistik.md.

Responsibility matrix — snow & precipitation (delineation vs. meteoswiss-mcp)

To avoid duplicating snow and precipitation data across the portfolio, responsibilities are split as follows. meteoswiss-mcp owns atmospheric precipitation and weather; swiss-environment-mcp (SLF domain) owns snow on the ground and avalanche danger.

Rule: snow on the ground and avalanche danger belong to swiss-environment-mcp (SLF); atmospheric precipitation (rain/snowfall as mm) plus weather, forecast, warnings and climate normals belong to meteoswiss-mcp. The SLF IMIS precipitation endpoint (RR_10MIN_SUM) is deliberately not wired up as a tool, so it does not duplicate MeteoSwiss. The snow/avalanche tools are live (see docs/probe-slf.md).

---

Testing

# Unit tests (no API keys or network required)
PYTHONPATH=src pytest tests/ -m "not live"

# Integration tests (requires live BAFU APIs)
PYTHONPATH=src pytest tests/ -m "live"

# Linting
ruff check src/

---

Contributing

See CONTRIBUTING.md (English) · CONTRIBUTING.de.md (German)

---

Security

Security policy and posture: SECURITY.md (English) · SECURITY.de.md (German). Full security architecture: docs/security.md.

---

Changelog

See CHANGELOG.md

---

License

MIT License — see LICENSE

Source data is subject to BAFU terms of use. Attribution to BAFU is required when using their data.

---

Author

Hayal Oezkan · github.com/malkreide

---

Credits & Related Projects

• Data: BAFU / Bundesamt für Umwelt · hydrodaten.admin.ch · naturgefahren.ch · opendata.swiss
• Protocol: Model Context Protocol – Anthropic / Linux Foundation
• Related:

Synergy example: "What was the air quality at Schulhaus Leutschenbach today – and how does it compare to the national NABEL average?"
→ zurich-opendata-mcp (OSTLUFT, local) + swiss-environment-mcp (NABEL, national)

• Portfolio: Swiss Public Data MCP Portfolio

Installation

Run via uv's uvx — no clone or manual install needed. Add to your MCP client config (mcpServers for Claude Desktop, Cursor and Windsurf; use a top-level servers key for VS Code in .vscode/mcp.json):

{
  "mcpServers": {
    "swiss-environment-mcp": {
      "command": "uvx",
      "args": [
        "swiss-environment-mcp"
      ]
    }
  }
}