Data.gov MCP. Search 300,000+ US government datasets by topic or agency.
Gemini Works with every AI agent you already use
…and any MCP-compatible client
Just plug in your AI agents and start using Vinkius.
Data.gov MCP Server connects your AI agent directly to 300,000+ US government datasets. You search, filter, and get full metadata on federal records—from climate and health to finance—without needing an API key.
Use the agent to find specific datasets by topic, organization, or tag, and download metadata formats like CSV, JSON, or GeoJSON.
It's a direct link to public US government data.
What your AI agents can do
Get dataset
Gets full metadata for a specific dataset by its ID or name, including license and resource list.
Get group
Gets details for a specific data topic group, such as 'agriculture' or 'health'.
Get group datasets
Gets all dataset titles, descriptions, and download links within a specific topic group.
Find specific US federal datasets using text search, filtering by organization, tags, or topic.
Retrieve full metadata for a single dataset, including its resource downloads, license, and data dictionary.
Get a list of major federal topics (e.g., agriculture, health, finance) and the datasets they contain.
List all federal organizations that publish open data and view their published datasets.
Find all datasets associated with a specific tag or category name.
Get a list of all supported file formats (CSV, JSON, XML, etc.) to narrow down data options.
Ask AI about this MCP
Supported MCP Clients
Gemini Waiting for input…
019d842dget dataset
Gets full metadata for a specific dataset by its ID or name, including license and resource list.
019d842dget group
Gets details for a specific data topic group, such as 'agriculture' or 'health'.
019d842dget group datasets
Gets all dataset titles, descriptions, and download links within a specific topic group.
019d842dget organization
Gets details for a specific federal organization, including its contact info and published data.
019d842dget organization datasets
Gets all datasets published by a specific organization.
019d842dget status
Checks the overall status and total counts of datasets, groups, and tags on the Data.gov API.
019d842dget tag
Gets details for a specific tag, including its name and count of associated datasets.
019d842dget tag datasets
Gets all datasets that share a specific tag or category.
019d842dlist groups
Lists all major data topic groups available on Data.gov, along with their descriptions and dataset counts.
019d842dlist organizations
Lists all federal organizations that publish data, along with their dataset counts and contact info.
019d842dlist resource formats
Lists all supported file formats (CSV, JSON, XML, etc.) available in the data resources.
019d842dlist tags
Lists all available tags used to categorize datasets, helping you discover common topics.
019d842dsearch datasets
Searches the entire catalog of US federal government open datasets using keywords and filters.
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 Data.gov, then connect any of our 4,700+ other servers whenever your AI needs more. One click, no limits.
- Use this MCP plus 4,700+ 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
What you can do with this MCP connector
Data.gov MCP Server connects your AI client directly to over 300,000 US government datasets. You search, filter, and get full metadata on federal records—from climate and health to finance—without needing an API key. Here's what you can do:
Search the entire dataset catalog
search_datasets lets you search the whole catalog of US federal government open datasets using keywords and filters. You can pinpoint specific datasets by topic, organization, or tag.
List all available data topics
list_groups pulls up all major data topic groups on Data.gov, giving you descriptions and dataset counts for topics like agriculture or health.
Identify publishing government agencies
list_organizations lists all federal organizations that publish data, giving you their dataset counts and contact info. You can then use get_organization_datasets to pull all the datasets published by a specific agency. get_organization gives you full details on any single organization, including its contact info and published data.
Filter by data tag or category
list_tags shows you every available tag used to categorize datasets, helping you find common topics. You can then use get_tag_datasets to pull all datasets that share a specific tag or category name. For a single tag, get_tag provides details, including the name and count of associated datasets.
Detail a specific dataset's metadata
get_dataset pulls full metadata for any specific dataset using its ID or name, including the license and a list of resources. If you need to know what a dataset contains, get_group_datasets pulls all dataset titles, descriptions, and download links within a specific topic group. You can also get all datasets published by an organization using get_organization_datasets.
List resource formats
list_resource_formats shows every supported file format—CSV, JSON, XML, etc.—available in the data resources so you can narrow down your data options.
How Data.gov MCP Works
- 1 Connect your AI agent to the Data.gov MCP Server.
- 2 Tell your agent the goal (e.g., 'Find datasets about climate change').
- 3 The agent uses the appropriate tool (like
search_datasets) and returns the results and metadata for you to review.
The bottom line is, your AI agent handles the complex querying and filtering of federal data, and you just get the usable answer.
Who Is Data.gov MCP For?
Researchers, investigative journalists, and data developers who need deep, reliable public records. If you're tired of manually jumping between government websites to piece together data, this is for you. You get access to 300,000+ federal datasets without needing an API key or knowing the specific internal catalog structure.
Find federal datasets for a thesis or project, checking for appropriate licenses and resource formats.
Discover government data to support a story, cross-referencing data published by different agencies like EPA or Census.
Integrate federal open data into an application or dashboard, using the detailed metadata and download links.
What Changes When You Connect
- Find everything you need with
search_datasets. Instead of running multiple manual searches, you tell your agent the topic—say, 'climate change'—and it pulls results from across the entire catalog, citing datasets like NOAA Climate Data and EPA emissions records. - Cross-reference sources easily. If you're writing a report on agriculture, you can first run
list_groupsto find the 'agriculture' group, then useget_group_datasetsto see every dataset published under that topic. - Avoid API key headaches. The server requires no authentication because all data is public domain. This means you can run complex queries using
search_datasetswithout worrying about rate limits or credentials. - Get granular control. If you know the source or the topic, use
get_organizationorget_group. For instance, checking data only from the Census Bureau viaget_organization_datasetskeeps your focus tight. - Format-aware data prep. Don't guess the file type. Run
list_resource_formatsfirst. Then, when you runget_dataset, you know exactly which format (JSON, Shapefile, CSV) you can expect to download. - Understand the scope. Use
get_statusto see the total count of datasets, organizations, and tags available. This gives you a quick sanity check on the sheer size of the catalog before you start querying.
Real-World Use Cases
Tracking multi-departmental health trends
A public health researcher needs data on both flu outbreaks (from CDC) and local school attendance records. Instead of visiting two separate government portals, they prompt the agent: 'Compare health data from CDC with education data.' The agent uses get_group_datasets (health) and get_group_datasets (education) and then runs search_datasets to pull the relevant records, solving the data silo problem.
Investigating economic disparities in a region
A journalist is writing about housing costs. They prompt the agent to 'Find all housing and finance data for the Midwest.' The agent first uses list_tags to find 'housing' and 'finance' tags, then uses get_tag_datasets for both, and finally uses search_datasets to filter the combined results by 'Midwest' geography, delivering a comprehensive data packet.
Building a data-rich visualization dashboard
A data developer needs to visualize all available data formats. They start by running list_resource_formats to confirm CSV, JSON, and Shapefile support. Then, they run get_organization_datasets for NASA to get a list of available resources, which they then feed into their visualization pipeline.
Researching historical climate impact
A student needs all data related to temperature and carbon emissions. They first run list_groups to find 'climate', then use get_group_datasets to see the resources. If they want to narrow it down, they use get_tag_datasets with the 'emissions' tag, solving the problem of information overload.
The Tradeoffs
Manual, sequential API calls
A user manually runs get_group for 'health', then copies the output list of datasets, and manually runs get_organization for CDC, then copies that list. This is slow, error-prone, and requires context switching.
→
Tell your agent to 'Find all health data published by the CDC.' The agent handles the sequence: it uses get_group (health), identifies the CDC, and then runs get_organization_datasets to pull everything in one go.
Searching by vague keywords
Typing 'data' into a general search field. This returns too many results and doesn't give structure or context.
→
Be specific. Instead of 'data', ask the agent to 'Search for education datasets published by the Department of Education.' The agent uses search_datasets with filters for both organization and topic.
Assuming a single data source
Only checking the USDA website for crop data, when the EPA might have related pollutant data.
→
Use list_tags to find related topics (e.g., 'agriculture' and 'pollutants'). Then, use get_tag_datasets for both to build a full, cross-agency picture.
When It Fits, When It Doesn't
Use this server if your goal is discovery: finding what data exists, what formats it comes in, or which agencies publish it. You need to know the scope before you can analyze the data. Don't use this if you just need to transform data (use a dedicated ETL tool) or if you need data from a private, paid source. If you need to know the total count of datasets, run get_status first. If you know the organization but not the topic, start with get_organization and look at its published datasets. If you only know the general subject, start with list_groups to narrow down your search scope.
Independent Platform Disclaimer: Vinkius is an independent platform and is not affiliated with, endorsed by, sponsored by, verified by, or otherwise authorized by Data.gov. 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 server provides 13 capabilities that interface natively with Claude, ChatGPT, Cursor, and any MCP client. No middleware. No custom integration required.
Available Capabilities
Finding federal data shouldn't require knowing the right agency or topic.
Today, pulling data from the US government is a nightmare of clicks. You start at Data.gov, then you click 'Health,' then you see a list of 10 groups. If you need flu data, you might have to check the CDC's page, then the EPA's page, and then the Census's page, just to compare rates. You spend more time navigating the site than analyzing the data.
With the Data.gov MCP Server, you just tell your agent, 'Compare flu data from CDC and EPA.' The agent handles the logic. It uses `search_datasets` across all these sources, pulls the metadata, and gives you a clean list of exactly what you need—no manual jumping required.
Data.gov MCP Server: Get specific records with `get_dataset`
Manually, you find a dataset title and then have to click into it, navigate sub-pages, and copy-paste the metadata to confirm the license and download format. It’s tedious, and you often miss the data dictionary.
Now, your agent runs `get_dataset` with the name. It returns all the metadata—the description, the license, the resource list, and the full data dictionary—in one go. You get the complete picture, instantly.
Common Questions About Data.gov MCP
Do I need an API key for the Data.gov MCP Server? (get_dataset) +
No, you don't need an API key. The entire Data.gov catalog is public domain, so your agent can access all the data without authentication issues.
How do I find all datasets related to a specific topic, like finance? (list_groups) +
Use list_groups first to see the official name (e.g., 'finance'). Then, use get_group_datasets to get a list of every dataset under that specific topic.
What is the difference between `search_datasets` and `get_dataset`? (search_datasets) +
search_datasets searches across the entire catalog using keywords and filters. get_dataset requires you to know the specific ID or name of the dataset you want.
Can I find all data published by a specific government agency? (get_organization_datasets) +
Yes. Use get_organization to get the agency's details, and then run get_organization_datasets to pull a list of every dataset they've published.
What formats can I download data in? (list_resource_formats) +
Run list_resource_formats to see a comprehensive list of all supported file types, including CSV, JSON, XML, and Shapefile. This helps you plan your data ingestion.
How can I find all datasets associated with a specific tag using `get_tag_datasets`? +
The get_tag_datasets tool retrieves all relevant datasets for a given tag. You pass the tag name, and it returns a list of titles, descriptions, organizations, and download links for every matching record.
If I know the organization name, should I use `get_organization` or `get_organization_datasets`? +
Use get_organization when you just need general details about the agency (like their mission or contact info). Use get_organization_datasets when you need a list of all datasets that agency has published.
What is the best way to discover all available topics before I start searching? (list_groups) +
list_groups provides a comprehensive list of all topic groups on Data.gov. You get the group names, descriptions, and even the current dataset count for each topic, which helps you plan your search.
Do I need an API key? +
No! Data.gov data is public domain and freely accessible. No authentication required.
How many datasets are available? +
Data.gov catalogs 300,000+ datasets from over 200 federal agencies including NASA, USDA, EPA, NOAA, Department of Education, Census Bureau and many more.
What organizations publish data? +
Over 200 federal agencies including NASA, USDA, EPA, NOAA, Department of Education, Census Bureau, Department of Transportation, FBI, CDC, FDA and many more.
What formats are available? +
Common formats: CSV, JSON, XML, Shapefile, GeoJSON, PDF, HTML, RDF, KML, ZIP. Use list_resource_formats to see all available formats.
Use it with your favorite AI tools
Connect this server to Cursor, Claude, VS Code, and more.
More in this category
Cohere
Access Cohere AI models via API — chat with Command models, generate embeddings, rerank documents and tokenize text from any AI agent.
Lucidworks Fusion (AI Search & Discovery)
Manage AI-powered search via Lucidworks Fusion — execute semantic queries, index documents, and monitor ML training jobs.
Vertex AI Search
Search across your enterprise data using Google's semantic search and generative AI grounding.
You might also like
Semantic Scholar
Search 200M+ academic papers with AI-powered TLDR summaries, influential citation tracking, and researcher profiles from the Allen Institute for AI.
GoTab
Manage your restaurant operations — menus, orders, and tabs via AI.
Banco do Brasil
Check balance, statements, pay Pix/Boleto, and manage your BB account via API.