Aareguru MCP Server
Provides Swiss Aare river swimming data including water temperature, flow rates, safety assessments, and forecasts. Enables AI assistants to answer questions about current conditions, compare cities, and provide safety recommendations based on official BAFU thresholds.
README
Aareguru MCP Server
MCP server for Swiss Aare river data, enabling AI assistants like Claude to answer questions about swimming conditions, water temperature, flow rates, and safety.
๐ Quick Start
Use directly from FastMCP Cloud (no installation needed):
Add it is as custom connector in Claude Desktop:

No authentication is needed.
Altnernatively, you can add the aareguru-mcp.mcpb file via option in Claude -> Settings -> Extensions. Or edit the Claude desktop config file directly:
{
"mcpServers": {
"aareguru": {
"url": "https://aareguru.fastmcp.app/mcp"
}
}
}
๐ธ Screenshots

๐ฏ Features
| Feature | Description |
|---|---|
| 6 MCP Tools | Temperature, flow, safety, forecasts, history |
| 4 MCP Resources | Direct data access via aareguru:// URIs |
| 3 MCP Prompts | Daily reports, spot comparisons, weekly trends |
| Rate Limiting | 100 req/min, 1000 req/hour protection against abuse |
| Metrics | Prometheus endpoint for monitoring and observability |
| Swiss German | Authentic temperature descriptions ("geil aber chli chalt") |
| BAFU Safety | Official flow danger levels and thresholds |
| Smart UX | Proactive safety warnings, alternative suggestions, seasonal context |
| 200+ Tests | 83% coverage, comprehensive test suite |
๐ ๏ธ Tools
| Tool | Description | Example Query |
|---|---|---|
get_current_temperature |
Water temperature with Swiss German text | "What's the Aare temperature?" |
get_current_conditions |
Full conditions (temp, flow, weather) | "How's the Aare looking today?" |
get_flow_danger_level |
Flow rate + BAFU safety assessment | "Is it safe to swim?" |
list_cities |
All monitored cities | "Which cities have data?" |
get_historical_data |
Temperature/flow history | "Show last 7 days" |
get_forecast |
Temperature/flow forecast | "Will it be warmer later?" |
BAFU Safety Thresholds
| Flow Rate | Level | Status |
|---|---|---|
| < 100 mยณ/s | Safe | Swimming OK |
| 100-220 mยณ/s | Moderate | Experienced swimmers only |
| 220-300 mยณ/s | Elevated | Caution advised |
| 300-430 mยณ/s | High | Dangerous |
| > 430 mยณ/s | Very High | Extremely dangerous |
๐ Resources
| URI | Description |
|---|---|
aareguru://cities |
List of all monitored cities |
aareguru://current/{city} |
Full current data for a city |
aareguru://today/{city} |
Minimal current data |
aareguru://widget |
Overview of all cities |
๐ฌ Prompts
| Prompt | Description |
|---|---|
daily_swimming_report |
Comprehensive daily report with conditions, safety, forecast, and recommendation |
compare_swimming_spots |
Compare all cities to find the best swimming spot today |
weekly_trend_analysis |
Analyze temperature and flow trends over the past week |
๐ป Local Installation
# Install uv and clone
curl -LsSf https://astral.sh/uv/install.sh | sh
git clone https://github.com/schlpbch/aareguru-mcp.git && cd aareguru-mcp
uv sync
# Run tests
uv run pytest
Claude Desktop (Local)
Edit ~/Library/Application Support/Claude/claude_desktop_config.json:
{
"mcpServers": {
"aareguru": {
"command": "uv",
"args": ["--directory", "/path/to/aareguru-mcp", "run", "aareguru-mcp"]
}
}
}
๐ณ Docker
cp .env.example .env
docker-compose up -d
curl http://localhost:8000/health
โ๏ธ Hosting
FastMCP Cloud (Recommended)
This server is deployed on FastMCP Cloud, a managed platform for MCP servers with zero-config deployment.
Features:
- โ Zero-Config Deployment - Connect GitHub repo, automatic deployment
- โ Serverless Scaling - Scale from 0 to millions of requests instantly
- โ
Git-Native CI/CD - Auto-deploy on push to
main, branch deployments for PRs - โ Built-in Security - OAuth support, token management, secure endpoints
- โ MCP Analytics - Request/response tracking, tool usage insights
- โ Free Tier - Available for personal servers
Deployment Steps:
- Sign in to fastmcp.cloud with GitHub
- Create Project and link your repository
- Deploy - Platform automatically clones, builds, and deploys
- Access - Get your unique URL (e.g.,
https://aareguru.fastmcp.app/mcp)
Configuration:
No special configuration needed! FastMCP Cloud auto-detects FastMCP servers. The server runs with:
- Health endpoint:
https://your-app.fastmcp.app/health - MCP endpoint:
https://your-app.fastmcp.app/mcp
Pricing:
- Free tier for personal projects
- Pay-as-you-go for teams (usage-based)
Alternative Hosting Options
FastMCP servers can be deployed to any Python-compatible cloud platform:
Container Platforms:
- Google Cloud Run
- AWS ECS/Fargate
- Azure Container Instances
PaaS Providers:
- Railway
- Render
- Vercel
Cloud VMs:
- AWS EC2
- Google Compute Engine
- Azure VMs
Deployment Pattern:
# For HTTP deployment, modify server to use HTTP transport
from fastmcp import FastMCP
mcp = FastMCP("aareguru")
# ... register tools ...
if __name__ == "__main__":
mcp.run(transport="sse") # Server-Sent Events for HTTP
Then containerize with Docker and deploy to your chosen platform.
๐ Monitoring & Observability
Prometheus Metrics
The server exposes Prometheus-compatible metrics at /metrics for monitoring:
Available Metrics:
aareguru_mcp_tool_calls_total- Counter of tool invocations by name and statusaareguru_mcp_tool_duration_seconds- Histogram of tool execution timesaareguru_mcp_api_requests_total- Counter of Aareguru API requestsaareguru_mcp_errors_total- Counter of errors by type and componentaareguru_mcp_active_requests- Gauge of currently active requests
Example:
curl http://localhost:8000/metrics
Rate Limiting
HTTP endpoints are protected with rate limiting:
- Default limits: 100 requests/minute, 1000 requests/hour
- Health endpoint: 60 requests/minute
- Headers: Rate limit info included in responses
- 429 responses: Automatic retry-after headers when limits exceeded
๐งช Development
uv run pytest # Run tests
uv run pytest --cov=aareguru_mcp # With coverage
uv run black src/ tests/ # Format
uv run ruff check src/ tests/ # Lint
๐ Project Structure
aareguru-mcp/
โโโ src/aareguru_mcp/ # Server, client, models, config
โโโ tests/ # 200 tests, 87% coverage
โโโ docs/ # API docs, testing, implementation
โโโ mcp_server.py # FastMCP CLI entry
โโโ pyproject.toml
๐ Data Attribution
Data from BAFU, Aare.guru, MeteoSchweiz, Meteotest.
Non-commercial use only - Contact: aaregurus@existenz.ch
๐ License
MIT License - See LICENSE
Built with โค๏ธ for the Swiss Aare swimming community
Recommended Servers
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.
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.
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.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
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.
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.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
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.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.