Meteostat MCP for AI. Model climate patterns from decades of historical weather.
Gemini Works with every AI agent you already use
…and any MCP-compatible client
How this MCP server connects to your AI agent
Meteostat connects your AI agent to one of the largest databases of historical weather and climate data. It handles everything from basic daily averages to complex, multi-decade climate normals.
Whether you need records for a specific station or must interpolate data for an arbitrary coordinate, this server gives you tools for hourly, daily, monthly observations across any location on Earth.
What AI agents can do with Meteostat Automation
Point daily
Gets historical weather statistics for an arbitrary point location on Earth, day by day.
Point hourly
Retrieves granular historical weather observations (like temperature and wind speed) for a specific coordinate, hour by hour.
Point monthly
Gets aggregated monthly weather statistics for any given point location.
Fetches extrapolated hourly, daily, or monthly weather records for any latitude and longitude on Earth.
Pulls recorded historical observations (hourly/daily/monthly) tied to specific, established weather station IDs.
Uses GPS coordinates and a radius to find the closest operational weather stations in the network.
Retrieves 30-year average temperature and precipitation data for both specific points and known station locations.
Looks up detailed information about a weather station, including its WMO or ICAO identifiers.
Ask an AI about this
Waiting for input…
What AI agents can do with Meteostat MCP Server: 10 Tools for Weather History
These tools let your AI agent query historical weather data at every level—from hourly point interpolation to multi-decade station normals.
Make your AI actually useful.
Add this MCP to Claude, Cursor, or Windsurf and your AI stops guessing. It gets real tools to look things up, take action, and handle the stuff you keep doing by hand.
Start using Meteostat on VinkiusPoint Daily
Gets historical weather statistics for an arbitrary point location on Earth, day by day.
Point Hourly
Retrieves granular historical weather observations (like temperature and wind speed)...
Point Monthly
Gets aggregated monthly weather statistics for any given point location.
Point Normals
Calculates the 30-year average climate normals (baseline data) for an arbitrary...
Stations Daily
Gets historical daily statistics from a known weather station ID; limited to 10...
Stations Hourly
Retrieves detailed hourly observations for an established station ID; limited to 30 days per request.
Stations Meta
Looks up descriptive details for a specific weather station using its unique identifier (WMO or ICAO).
Stations Monthly
Gets summarized historical monthly statistics from an established station ID.
Stations Nearby
Searches and lists all available weather stations within a specified radius of given...
Stations Normals
Retrieves the 30-year average climate normals (baseline data) for an established...
Security and governance baked right in.
Pick your AI client below to get set up. Just create a Vinkius account, subscribe, and you're instantly up and running. We handle the entire backend infrastructure, delivering out-of-the-box support for HTTPS Streamable, SSE, and OAuth2—zero messy routing required.
Claude AI
Open Claude Settings
Go to claude.ai, click your profile icon, then navigate to Customize → Connectors.
Add Custom Connector
Click the "+" button and select Add custom connector. Paste your Vinkius endpoint URL:
https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp
Replace [YOUR_TOKEN_HERE] with your token
from cloud.vinkius.com. For OAuth-protected servers, expand
Advanced settings to add credentials.
Start a conversation
Open a new chat. The Meteostat integration is available immediately — no restart needed.
Choose How to Get Started
Build a custom MCP for your own tools, or connect a ready-made integration from our catalog.
Build Your Own
Turn any API into an MCP. Import a spec, define Agent Skills, or deploy with MCPFusion.
- Import from OpenAPI, Swagger, or YAML specs
- Create Agent Skills with progressive disclosure
- Deploy to edge with MCPFusion framework
- Built in DLP, auth, and compliance on every call
- Real time usage dashboard and cost metering
- Publish to catalog or keep private
Make Your AI Do More
Start with Meteostat, then connect any of our 5,100+ other servers whenever your AI needs more. One click, no limits.
- Use this MCP plus 5,100+ others, all in one place
- Add new capabilities to your AI anytime you want
- Every connection is secured and compliant automatically
- Track usage and costs across all your servers
- Works with Claude, ChatGPT, Cursor, and more
- New servers added to the catalog every week
Independent Platform Disclaimer: Vinkius is an independent platform and is not affiliated with, endorsed by, sponsored by, verified by, or otherwise authorized by Meteostat. All third-party trademarks, logos, and brand names are the property of their respective owners. Their use on this website is strictly for informational purposes to identify service compatibility and interoperability.
VINKIUS INFRASTRUCTURE
Cloud Hosted
Managed infra
V8 Isolated
Sandboxed per request
Zero-Trust Proxy
No stored credentials
DLP Enforced
Policy on every call
GDPR Compliant
EU data residency
Token Compression
~60% cost reduction
Built on the Model Context Protocol (MCP) for Claude, ChatGPT, Cursor, and more
The Model Context Protocol standardizes how applications expose capabilities to LLMs. Instead of operating in isolation, your AI gains direct access to external platforms, live data, and real-world actions through secure, standardized connections.
This connection provides 10 powerful capabilities that interface natively with Claude, ChatGPT, Cursor, and other compatible AI platforms. No middleware. No custom integration required.
Trying to find weather data always feels like a scavenger hunt., Solved with Vinkius AI Gateway
Right now, if you need historical climate context, you usually have to juggle three separate systems: first, finding the closest official station ID; second, figuring out if that station has hourly or monthly records; and third, manually calling different endpoints for every time scale (daily vs. monthly) just to get a complete picture.
With Meteostat, your agent handles the complexity. You just ask, 'What was the climate like in Chicago in 2015?' The server determines if it needs to use `stations_nearby`, then runs `stations_monthly` and packages the full context into one clean answer.
Meteostat MCP Server: Get data for any point or station.
The biggest time sink used to be figuring out if you needed a 'point' calculation (interpolation) when no station existed, or if you should use the recorded `stations_*` tools. This distinction was always confusing and prone to manual error.
Now, your agent manages that decision for you. You tell it the goal; it selects whether to interpolate using `point_daily` or pull verified data from a known location using `stations_daily`. It’s cleaner, faster, and far less error-prone.
What your AI can actually do with this
You need historical weather data for your project? This Meteostat server connects your AI agent to one of the biggest databases out there. Forget building complex scrapers or juggling multiple API endpoints; you just let your agent use specific tools to pull precise meteorological records for any time scale and location type.
Querying Specific GPS Points: When You Don't Know a Station ID
If you only have coordinates—latitude and longitude—and no known weather station nearby, you can still get solid data. The server gives your agent tools that extrapolate historical observations for that exact spot on Earth. To pull hourly records, use point_hourly to retrieve granular measurements like temperature or wind speed hour by hour.
If you need the whole day summarized, point_daily gets those full daily statistics. For a broader view of seasonality, point_monthly pulls aggregated monthly weather stats for any given point.
Accessing Established Weather Stations: Real Records Only
When you know an established station ID (like one from the national network), your agent reads actual recorded measurements—no guessing involved. If you're looking at detailed records, stations_hourly retrieves hourly observations for that specific station; just remember it’s limited to thirty days per request. For a full day summary, use stations_daily, keeping in mind that this tool caps requests at ten years of data.
Need the monthly rundown from a known station? stations_monthly gives you those summarized historical statistics.
Finding Stations and Checking Metadata
You don't always know the ID you need. The server helps you figure it out first. If your target location is unknown, use stations_nearby. You just give it GPS coordinates and a radius, and it spits back a list of all operational weather stations within that zone. Once you have an ID, you can check its background using stations_meta; this tool looks up detailed descriptive info about the station, including whether it uses WMO or ICAO identifiers.
Calculating Climate Normals: Long-Term Baselines
For long-term research, your agent needs baseline data. The server gives you two ways to calculate the 30-year average climate normals. You can run point_normals on an arbitrary coordinate to get a general area's baseline temperature and precipitation figures. Alternatively, if you want the official historical normal for a known station, use stations_normals with that station’s ID.
Summary of Capabilities
- Interpolation: Get extrapolated hourly, daily, or monthly weather records using coordinates via
point_hourly,point_daily, andpoint_monthly. - Direct Station Query: Pull recorded historical data (hourly/daily/monthly) tied to specific station IDs through
stations_hourly,stations_daily, andstations_monthly. - Station Discovery: Find the closest monitoring sites using GPS coordinates via
stations_nearby, then check details withstations_meta. - Climate Baseline: Determine 30-year averages—either for a specific point (
point_normals) or an established station (stations_normals).
Your AI agent uses these tools to handle everything from basic daily averages to complex, multi-decade climate normals. You just tell it what you need.
019e5d35-58ac-734f-be63-e75d9831e053 
Here's how it actually works
The bottom line is: you write a prompt, your AI agent figures out which historical dataset to hit, runs it for you, and gives you clean numbers.
Subscribe to the Meteostat Server and input your API Key into your AI client.
Ask your agent for specific historical data (e.g., 'What was the average temp in Paris on June 12, 2019?').
The agent selects the correct tool (point_daily, stations_nearby, etc.), executes it with parameters, and returns the structured weather data.
Who is this actually for?
Environmental consultants, climate researchers, or logistics managers use this. You're the person who wakes up needing data that spans years—not just today’s forecast. If your job requires comparing current conditions to a 30-year baseline, you need this server.
Uses stations_normals and point_daily to build machine learning models that predict regional climate shifts over decades.
Checks historical weather using stations_nearby before planning major, multi-day routes through unfamiliar regions.
Pulling specific monthly averages (stations_monthly) to determine if a site's current water runoff levels are within typical 30-year norms.
What Changes When You Connect
You can pull hourly records for a single point using point_hourly, letting you track micro-changes at any exact latitude/longitude, perfect for localized environmental models. This avoids the need to guess which station is closest and provides continuous data flow.
When you're planning an event or construction site far from existing infrastructure, use stations_nearby first. It maps out all local stations so you know exactly what kind of historical context data is available for your area.
Comparing current conditions to a 30-year average is fast. Both point_normals and stations_normals let you establish a reliable climate baseline with one tool call, which is critical for impact studies.
The server separates station records from point interpolation. If you need recorded data (e.g., 'What did Station 123 actually measure?'), use the stations_* tools. If you just have coordinates, stick to the point_* suite.
Need more context on a specific site? Use stations_meta to pull WMO or ICAO IDs before querying historical data. This ensures your agent is using the correct identifier for deep lookups.
See it in action
Analyzing wildfire risk over a decade.
A forestry consultant needs to know if current drought levels are anomalous. They first use stations_nearby to find stations near the burn zone, then run stations_daily and compare the results against the long-term baseline provided by stations_normals for that region.
Determining optimal placement for a new solar farm.
A developer doesn't have an existing station ID. They use point_daily on multiple candidate coordinates to model the average annual solar irradiance and temperature fluctuations, ensuring they pick the site with the most consistent historical output.
Assessing airport viability for expansion.
The planning team needs reliable historical wind data. They use stations_hourly on the nearest major airport station ID to gather granular flight pattern data, which is much more precise than general point interpolation.
Building a global climate change model.
A research team pulls long-term monthly averages (point_monthly) from several different coordinates across the globe. This allows them to build models that account for regional variability without needing data from every single existing station.
The honest tradeoffs
Confusing point interpolation with network data
Running point_daily when you actually need the recorded measurements from a nearby weather tower.
Don't use general coordinates if a known station exists. First, run stations_nearby to get the ID, then pull the specific history using stations_daily or stations_hourly for accurate network data.
Trying to model real-time weather
Asking the agent what the temperature is 'right now' and expecting a historical API tool to respond.
This server only handles historical records. For live data, you need a different streaming service. Use point_daily or similar tools only when your question includes specific dates.
Forgetting the time scale difference
Asking for yearly averages using stations_hourly, which will fail or return incomplete data.
Always match the granularity. If you need monthly summaries, use stations_monthly. For full-year analysis, consider combining multiple point_monthly calls.
When It Fits, When It Doesn't
Use this server if your core requirement involves historical context: comparing current conditions to past trends, or modeling climate shifts over months or years. The tools are structured around two axes: Location Scope (Is it an arbitrary point? Or a known station?) and Time Granularity (Hourly, Daily, Monthly).
* Use this if: You need to model long-term trends, compare climate normals (stations_normals) against recent data, or calculate the average conditions for a location that doesn't have an official weather tower. The point_* tools are your best bet here.
* Don’t use this if: Your goal is real-time monitoring (you need live feeds), or if you only care about the most recent 24 hours and don't need a historical comparison. In those cases, check for specialized streaming APIs instead.
Questions you might have
How do I get historical daily weather for a random coordinate? +
Use the point_daily tool. This tool extrapolates data across the entire globe based on coordinates, making it perfect when you don't have an official station ID nearby.
Can I get 30-year climate normals for a specific weather station? +
Yes, use stations_normals. This tool reads the established long-term average data (the 'normal') directly from a known station ID, providing vital context against recent measurements.
What's the difference between `point_daily` and `stations_daily`? +
Point_daily interpolates data for coordinates you provide; it assumes weather patterns based on surrounding stations. Stations_daily reads actual, recorded measurements taken by a specific, established station ID.
Which tool should I use to find nearby weather monitoring sites? +
Start with the stations_nearby tool. Give it your GPS coordinates and radius, and it will return a list of available stations you can then query using their IDs.
How do I get hourly data for a point location? +
Use point_hourly. This retrieves granular historical observations (temp, wind, etc.) for an exact coordinate over time. Remember this tool has specific limits per request.
What identifiers should I provide to `stations_meta` if I need detailed information about a weather station? +
The tool requires you to specify at least one identifier. You must include the station's ID, WMO number, or ICAO code in your request payload. This process lets you verify which stations are available before pulling any time-series data.
If I only need annual summaries, is `stations_monthly` better than getting and processing daily records with `stations_daily`? +
Yes. Use stations_monthly directly for streamlined access to seasonal data. It aggregates the required statistics at a monthly level, saving you from fetching and having to process dozens of individual day entries.
How does `point_normals` calculate climate averages, and what is the required input for global location data? +
The tool calculates 30-year average climate normals based on a precise latitude/longitude coordinate. You must provide exact decimal coordinates in your request to define the single geographic point you are analyzing.
How can I find weather data for a location that doesn't have a specific weather station? +
You can use the point_hourly or point_daily tools. These tools use interpolation to calculate weather data for any geographic coordinate (latitude/longitude) by combining data from surrounding stations.
What is the difference between historical data and climate normals? +
Historical tools like stations_daily provide actual observations for specific dates. The stations_normals tool provides long-term statistical averages (usually over 30 years), which represent the 'typical' weather for a location.
Can I get weather data in Fahrenheit instead of Celsius? +
Yes. Most tools, such as stations_hourly and point_daily, include an optional units parameter. You can set this to 'imperial' to receive data in Fahrenheit and other non-metric units.
We've already built the connector for Meteostat. Just plug in your AI agents and start using Vinkius.
No hosting. No infrastructure. No complex setup.
All 10 tools are live and waiting.
You're up and running in seconds.
Gemini Vinkius gives your AI agents access to the full catalog of app connectors, all fully managed, secure, and enterprise-ready. One subscription, every tool you need.
Built, hosted, and secured by Vinkius. You just connect and go.