NCEI Climate Data MCP for AI. Retrieve NOAA's historical weather records via natural language.
Gemini Works with every AI agent you already use
…and any MCP-compatible client
Connect to your AI in seconds.
NCEI Climate Data Online (NOAA Archive) gives your AI client direct access to NOAA's National Centers for Environmental Information archive.
It lets you search, list, and retrieve decades of historical weather records—from temperature averages (TAVG) to precipitation totals (PRCP)—using natural language queries.
What your AI can do
List datacategories
Lists high-level groupings of available climate datasets to narrow down your research focus.
Get data
Fetches actual climate observations (annual/monthly) after defining location, variable, and time range.
List datatypes
Lists the available climate variables and their codes (like TAVG or PRCP) that you can query.
Use list_datasets and list_datacategories to see all primary data archives NOAA maintains.
The agent runs list_stations to identify exact observing platforms worldwide, giving you the source ID needed for queries.
You can use list_locations or list_locationcategories to narrow your search down to countries, states, or specific zip codes.
Run list_datatypes to confirm the exact code for measurements like precipitation (PRCP) or average temp (TAVG).
Use search_data to find relevant datasets by specifying both a date range and a geographical area.
The agent executes get_data using all the metadata gathered (station ID, data type, dates) to fetch the final numbers.
Ask an AI about this
Waiting for input…
NCEI Climate Data Online (NOAA Archive): 10 Tools for Weather Records
This collection of tools allows your AI client to perform complex metadata discovery and structured data retrieval across NOAA's full archive.
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 NCEI Climate Data Online (NOAA Archive) on VinkiusList Datacategories
Lists high-level groupings of available climate datasets to narrow down your research focus.
Get Data
Fetches actual climate observations (annual/monthly) after defining location...
List Datatypes
Lists the available climate variables and their codes (like TAVG or PRCP) that you...
List Datasets
Finds information about specific, pre-packaged NCEI datasets (e.g., Global Summary...
List Locationcategories
Lists groupings of geopolitical areas, such as 'Countries' or 'States', to organize...
List Locations
Provides a list of specific geopolitical entities or bounding coordinates for data retrieval.
List Stations
Lists every weather observing platform (station) available in the NOAA network by ID and name.
Search Data
Discovers relevant climate data points based on combined temporal and spatial...
Search Datasets
Finds available datasets by matching them against specific time periods or locations.
Get Service Data
Accesses subset data in multiple formats when the standard retrieval method isn't...
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 NCEI Climate Data 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 NCEI Climate Data Online (NOAA Archive), 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 NOAA NCEI. 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
Works with 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.
Figuring out which climate dataset is even available shouldn't feel like writing a thesis on NOAA documentation.
Right now, if your agent needs historical rainfall data for the Gulf Coast, you have to jump through hoops. You check one page for location IDs; another page for variable codes (PRCP); and a third section just says 'Check here for available datasets.' It’s three separate research tasks that requires copy-pasting and cross-referencing across multiple manual tabs.
With this MCP server, you give your agent the task. The system uses `list_locationcategories` to scope out the region, then checks `list_datasets` to confirm NOAA has a record for that area. It gathers all the metadata in one go and presents it back to you: 'We found three options—A, B, or C. Which do you want?'
The NCEI Climate Data MCP Server lets you pull precise historical data with `get_data`.
Before this server, retrieving the actual numbers involved a painful process of compiling location IDs, variable codes, and date ranges into one massive, complex API call that frequently failed or required manual cleanup. You'd spend hours just validating the parameters before you even saw the data.
Now, your agent handles all that validation automatically. It collects everything it needs—the station ID from `list_stations`, the variable code from `list_datatypes`—and executes a single, clean call to `get_data`. You get structured observations right away.
What your AI can actually do with this
You're connecting your AI client straight into the NOAA Climate Data Online archive. It’s basically NOAA's National Centers for Environmental Information (NCEI) data dump, and it lets you query decades of global historical weather records using standardized structures. You don't have to manually dig through APIs; your agent handles the whole process.
To figure out what data you need, you start broad. If you want a general idea of what kind of archives NOAA keeps—like daily summaries or major global averages—you use list_datasets and list_datacategories. These tools show you all the primary groupings of climate datasets available in the archive.
If you're trying to narrow down your research focus, you can first check list_locationcategories. This shows you high-level ways to group areas, like 'Countries' or 'States.' To get specific geography, you run list_locations, which gives you a list of geopolitical entities or precise bounding coordinates needed for any query. You can also use list_locationcategories to filter down your search.
When it comes to the actual weather variables—you know, if you need average temperature (TAVG), precipitation totals (PRCP), or maximum temperatures (TMAX)—you don't guess. You run list_datatypes. This tells you the exact variable codes you need to use in your queries.
To find out which specific weather stations are even part of this network, you check list_stations. That tool lists every observing platform across the NOAA grid by both its ID and its name. You can also use list_datasets to find information on pre-packaged NCEI datasets, like the Global Summary of the Month.
When you need to search for data that fits specific parameters—say, temperature records in Florida between 1980 and 2000—you run search_data. This function discovers relevant climate points by matching both a date range and a geographical area. For finding datasets based on time or location alone, use search_datasets.
If the standard data retrieval isn't enough for what you need, you have two ways to pull numbers. First, you run get_data. This fetches actual climate observations—whether they’re annual averages or monthly totals—after you’ve defined the specific location, variable code, and time window. If that function falls short, get_service_data lets your agent access subset data in multiple formats when a standard retrieval method won't cut it.
To pinpoint exact stations globally, you run list_stations. To organize your search by region, you use list_locations or list_locationcategories. If you need to know what variables are available—like TAVG or PRCP—you check list_datatypes. You can see all the primary archives NOAA maintains using list_datasets and list_datacategories, and if you're just looking for general data points matching a time and place, run search_data.
To find pre-packaged datasets by location or date, use search_datasets. When you finally have every piece of metadata—the station ID, the variable code, and the dates—you execute the final query using either get_data or get_service_data to pull those raw historical observations.
019e38c6-6999-71ea-81b3-56be6b17ff6d 
Here's how it actually works
The bottom line is: Your AI client takes your natural language request and converts it into a precise multi-step sequence of API calls to pull NOAA's raw data.
Subscribe to this server and get a free API Token from the NOAA NCEI portal. You must provide this token to your AI client.
Instruct your agent on the task (e.g., 'Find precipitation data for Miami in 2015'). The agent will first use list_datatypes and list_locations to gather necessary codes.
The agent then passes all collected parameters—the location ID, the variable code, and the date range—to get_data. You get back structured climate observations.
Who is this actually for?
Researchers, environmental consultants, and developers who deal in historical climate or geo-spatial data. You are the person stuck manually navigating government websites and writing brittle API wrappers just to check if a location saw record rainfall last year. This server lets you ask complex questions directly.
You use this to pull station metadata (list_stations) and historical weather data for impact reports, confirming local environmental changes.
You run multi-step queries, chaining search_datasets with get_data, to build time series models without leaving your agent environment.
You query specific variables (like TAVG or PRCP) across decades and multiple global locations for academic papers, letting the AI handle the data retrieval logistics.
What Changes When You Connect
Stop manually building API calls. Instead of figuring out which endpoints handle location vs. variable data, let your AI client use list_locations and list_datatypes to collect all the necessary metadata first. It handles the complex pre-work for you.
Get specific climate metrics instantly. When you need average temperature (TAVG) or precipitation (PRCP), your agent doesn't guess; it uses get_data with confirmed NOAA codes, guaranteeing you pull accurate observations.
Filter massive datasets efficiently. If you only care about the 1980s in Florida, use search_data. This tool combines time and space parameters to cut out years of irrelevant data immediately.
Track down exact physical sources. Need to know which weather station recorded a reading? Run list_stations to get the platform's official ID before you even try to pull the data with get_data. It gives you source accountability.
Avoid dataset confusion. Instead of sifting through NOAA documentation, use list_datasets to see clear examples like 'Daily Summaries (GHCND)' and pick exactly what you need for your analysis.
See it in action
Assessing historical drought risk in the Midwest
A farmer needs to know if 2018 was historically dry. They ask their agent: 'What was the average precipitation (PRCP) for Iowa between 2015 and 2020?' The agent runs list_datatypes to confirm PRCP, then uses list_locations for Iowa's coordinates, and finally calls get_data with all three pieces of info. Result: A clean time series showing the rainfall totals.
Validating a site’s historical operational parameters
An environmental consultant needs to verify that a construction site near Heathrow airport has historically high levels of frost damage. They use list_stations to find the official station ID, then run search_data for 'TMIN' (Minimum Temperature) over 30 years. The agent returns all records, allowing them to confirm the historical freeze/thaw cycles.
Comparing regional temperature shifts across decades
A climate scientist wants to compare average temperatures between two distinct regions (e.g., Miami vs. Seattle). The agent uses list_locations twice, gets both location IDs, and then runs a single get_data call that compares TAVG for the same time period across both locations. Problem solved in minutes.
Checking available data types for an obscure region
A developer needs to know if NOAA tracks snow depth (SNOW) for a small, remote mountain town. Instead of reading the docs, they use list_datatypes first and then cross-reference that list with search_datasets using the town's location ID. The agent confirms availability or flags it as missing.
The honest tradeoffs
Asking for data without defining variables
The user simply asks, 'Get me the climate data for Paris.' This is too vague. The agent doesn't know if they want temperature, rain, or wind speed.
You must first run list_datatypes to confirm the variable code (e.g., TAVG). Then use that confirmed code along with get_data and a location ID for a complete query.
Treating it like a simple web search
Copy-pasting parameters into a single, massive prompt: 'Data for Paris, 2015, TAVG, PRCP.' The agent might get confused about which tool takes precedence.
Always use the metadata tools first. Run list_locations to validate the location ID, then run search_data to confirm the date range works with that spot.
Skipping the dataset discovery phase
Trying to pull data directly using only a location and dates without checking if NOAA has packaged it correctly in an archive.
Start by running list_datasets. This tells you which pre-built, reliable archives exist for your time period before attempting raw retrieval.
When It Fits, When It Doesn't
Use this server if your goal is scientific rigor and deep historical data access. If you need to model climate trends or compare metrics across years, this is the tool. You must know that this only provides highly structured, archival NOAA numbers—it does not provide real-time weather updates (for those, look at live weather API tools). Don't use it if your primary goal is just a simple check: 'What's the temperature right now?' That requires a different service. If you're struggling to map out a complex query, don't panic; run list_datatypes first. It confirms what data points exist before you waste time trying to retrieve them.
Questions you might have
How do I find all available data types using list_datatypes? +
You run the list_datatypes tool. This returns a comprehensive list of NOAA variable codes and their descriptions, like TAVG (Average Temperature) or PRCP (Precipitation). It's your starting point for defining metrics.
What is the difference between search_data and get_data? +
search_data finds if data exists by matching general parameters (time/space). get_data retrieves the actual, structured observations once you've confirmed that the necessary metadata and dataset are in place.
Do I need to use list_stations before running get_data? +
Yes. To ensure data accuracy and accountability, you should run list_stations first. This confirms the specific platform ID (e.g., GHCND:UKM00003772) that recorded the numbers you want.
Can I find climate data for a location not listed? +
If your exact area isn't in NOAA’s primary list, use list_locations to see if it falls under a general bounding box or region. If so, you can use that broader ID instead.
How do I get started with this server? What token do I need before using tools like `get_data`? +
You must request a free API Token from the NOAA NCEI portal. This token authorizes your AI client to access and query the historical data stream, so you'll need it for every call.
If I don't know if a location is a Country or State, how do I use `list_locationcategories`? +
The tool returns groupings of similar locations, like 'Countries' or 'States'. Run this first to understand the data hierarchy before using list_locations to pinpoint specific geopolitical entities.
What are the maximum time ranges I can retrieve when running `get_data`? +
The limits depend on the dataset. Annual or monthly data usually restricts you to a 10-year range, while other climate observation types might be limited to just one year.
How can I use `get_service_data` to ensure the output is in multiple formats? +
This tool lets your agent access subset data and specify various output formats. It pulls the same information structured for different downstream applications, which is really useful.
How do I find the specific ID for a weather station in a certain city? +
You can use the list_stations tool and provide a locationid. To find the correct location ID first, use the list_locations tool to search by city or state name.
What is the difference between a Data Category and a Data Type? +
Data Categories (retrieved via list_datacategories) are broad groups like 'Temperature' or 'Precipitation'. Data Types (retrieved via list_datatypes) are specific codes like 'TMAX' (Maximum temperature) or 'PRCP' (Precipitation amount).
Can I see what datasets are available for a specific date range? +
Yes, the list_datasets tool accepts startdate and enddate parameters. This allows you to filter the archive for datasets that have coverage during your period of interest.
We've already built the connector for NCEI Climate Data. 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.