OpenDataSUS MCP for AI. Query Brazilian Public Health Data Directly.

Q: How do I find all possible datasets using packagelist?

You run packagelist. This tool provides a comprehensive list of every dataset name available on the portal. It's your starting point for seeing what data exists.

Q: What is the difference between packagesearch and datastoresearch?

packagesearch finds the dataset (the container). datastoresearch runs the query against the actual rows of data inside that dataset, pulling out the results you need.

Q: Can I check a resource's columns using resourceshow?

Yes. Use resourceshow to get technical metadata for any specific file linked to a package. This tells you exactly what columns (like 'municipio') and data types are available.

Q: If I need to know who provided a dataset, should I use organizationlist first?

Yes, organizationlist provides a definitive list of all data providers. You can identify the source organization before running any searches or analyzing specific packages.

Q: What detailed information do I get when I run packageshow?

packageshow delivers the full metadata package for a dataset. This includes provenance, licensing terms, and structural details—more than just a simple description.

Claude

ChatGPT

Cursor

Gemini

Windsurf

VS Code

JetBrains

Vercel

See Vinkius in Action

Works with every AI agent you already use

…and any MCP-compatible client

Connect to your AI in seconds.

OpenDataSUS connects your AI client directly to Brazil’s official public health data (SUS) portal. It lets you search, filter, and pull actual rows from massive datasets—like COVID-19 vaccination records or epidemiological stats—without having to download a single file.

Use its tools to explore everything the Ministry of Health has published in natural language.

What your AI can do

Datastore search

Filters and retrieves specific rows of data from a given resource.

Group list

Lists all high-level categories used to group datasets on the portal.

Organization list

Provides a list of official departments or organizations that publish data.

+ 5 more capabilities included

Discover all available datasets

Use package_list to get a complete catalog of every dataset name on the OpenDataSUS portal.

Search for specific data packages

Filter and search for datasets using keywords or criteria with package_search.

Identify data sources and providers

List all official organizations that provide health data, like the Ministério da Saúde, using organization_list.

Retrieve full dataset schema

Get detailed metadata for a specific package or resource using package_show or resource_show, showing its structure and provenance.

Filter and retrieve raw data rows

Use datastore_search to query the actual content of a resource, pulling filtered table data directly into your conversation.

Ask an AI about this

Included with Plan

Waiting for input…

AI Agent

OpenDataSUS: 8 Tools for Public Health Analytics

These eight tools let you systematically discover, inspect, and query every aspect of the Brazilian public health data available on the OpenDataSUS portal.

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 OpenDataSUS on Vinkius

Datastore Search

Filters and retrieves specific rows of data from a given resource.

Group List

Lists all high-level categories used to group datasets on the portal.

Organization List

Provides a list of official departments or organizations that publish data.

Package List

Lists every single dataset package available across the OpenDataSUS portal.

Package Search

Searches for specific datasets by name or description criteria.

Package Show

Retrieves detailed metadata about a chosen dataset, including its purpose and resources.

Resource Show

Gets the technical metadata for an individual data file (like CSV), detailing columns and format.

Tag List

Lists all keywords or tags used across datasets, helping you scope a general topic.

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 OpenDataSUS integration is available immediately — no restart needed.

Antigravity

Configure Agent Environment

Open your Antigravity agent's workspace configuration or mcp-servers.json file.

Bind the Endpoint

Add the Vinkius endpoint URL to your agent's MCP connections list:

"mcp_servers": {
  "opendatasus": {
    "serverUrl": "https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp"
  }
}

Provide your secure token in place of [YOUR_TOKEN_HERE] to ensure your agent requests are authenticated.

Execute

Start your Antigravity session. The agent will autonomously discover and utilize the OpenDataSUS tools with full Vinkius guardrails applied.

VS Code Copilot

⚡

One-Click Install (Recommended)

In your Vinkius Dashboard, simply click the Add to VS Code button for this server. We'll automatically configure your local workspace.

Or configure manually

Open MCP Settings

Open VS Code, press Ctrl/Cmd + Shift + P, and search for GitHub Copilot: MCP Servers.

Add Server Config

Add the Vinkius endpoint configuration to your mcp-servers.json file:

"opendatasus": {
  "url": "https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp"
}

Ensure you replace [YOUR_TOKEN_HERE] with your token from cloud.vinkius.com.

LangChain

Install Dependencies

Install the LangChain MCP adapters for your environment:

pip install langchain-mcp-adapters

Connect the Server

Use the SSEClient in LangChain to connect to the Vinkius managed endpoint:

from langchain_mcp_adapters.client import SSEClient

# Connect to Vinkius
client = SSEClient(url="https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp")
tools = client.get_tools()

CrewAI

Define the Tool

Load the Vinkius MCP tools into your CrewAI agents:

from crewai import Agent
from mcp_crewai import MCPTool

# Connect securely to Vinkius
vinkius_tools = MCPTool(url="https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp")

# Assign to Agent
researcher = Agent(
    role='Data Researcher',
    tools=vinkius_tools.get_all()
)

Execute Task

Run your CrewAI process. The agent will autonomously route tasks to the Vinkius managed server.

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

Start building

Make Your AI Do More

Start with OpenDataSUS, 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 OpenDataSUS. 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

Your data is protected. See how we built it.

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 8 powerful capabilities that interface natively with Claude, ChatGPT, Cursor, and other compatible AI platforms. No middleware. No custom integration required.

Finding health statistics shouldn't require manual portal hopping.

Today, finding a simple statistic means navigating multiple official government portals. You search the main site, hit dead ends, find a link to an Excel sheet buried in a PDF, and then you have to manually download that file just to check the headers. It's copy-paste hell.

With OpenDataSUS, your agent handles it all. You ask: 'What were the confirmed cases for this region?' The server runs `package_search` to find the right dataset, confirms its structure with `resource_show`, and then uses `datastore_search` to deliver only the relevant cells—no zip file needed.

OpenDataSUS MCP Server: Structured Health Data Discovery

Manual research requires you to remember which department (organization) holds what data, and whether that dataset is even structured for your query. You spend time vetting the source before you get to the numbers.

Now, simply ask. The MCP Server manages the complexity of the CKAN API behind the scenes. It lets you treat the entire government data portal like a single search engine, giving you direct access to validated metrics.

Support 24/7 support@vinkius.com ↗

Security Vinkius Trust Center ↗

SLA Service Level Agreement ↗

Report Listing Send Report ↗

What your AI can actually do with this

OpenDataSUS connects your AI client straight into Brazil’s official public health data (SUS) portal. You get direct access to massive datasets—think COVID-19 vaccination records or deep epidemiological stats—and you never have to download a single file. Your agent uses these tools to explore everything the Ministry of Health has published, right in your chat window.

Getting Started: Discovering What’s Available

You gotta know what data exists before you can use it. You start by figuring out the scope. Use package_list when you need a complete catalog; this dumps every single dataset name available across the entire OpenDataSUS portal. If that list is too big, you narrow your focus first. You can check which organizations provided the info using organization_list, listing official departments like the Ministério da Saúde.

Or, if you're looking for a general subject—say, 'vaccinations' or 'COVID-19'—you run tag_list to pull all available keywords and tags that help scope your topic.

If you know what you’re looking for, but not the exact name, use package_search. You feed it keywords or criteria, and it filters down the dataset list. Once you have a potential package, you run package_show to pull detailed metadata on that specific dataset—you'll see its stated purpose and what resources it contains.

This is your high-level overview.

Deep Diving Into Data Structure

Knowing a dataset exists isn’t enough; you gotta know what columns it has. If package_show points to a resource file (like a CSV), use resource_show. This tool gets the technical metadata for that individual data file, detailing every column name and its format before your agent queries it. It's how you confirm if a field is a date, an integer, or a string.

Querying the Raw Data

This is where the magic happens. You use datastore_search to query the actual content of a resource. Instead of just getting metadata, this tool pulls filtered table data directly into your conversation. You tell your agent exactly which columns and what rows you need, and it returns the raw data—the full result set—ready for analysis right in the chat.

This capability lets you treat the dataset like an active database connection.

The Workflow Summary

Your typical workflow runs through these steps:

Scope: You run tag_list or organization_list to narrow down general topics or providers.
Search/Filter: You use package_search to pinpoint the right dataset.
Inspect: You run package_show to understand the package’s scope, and then resource_show to confirm the technical schema (columns and data types) of the underlying file.
Extract: Finally, you execute datastore_search, telling the agent precisely what rows and columns to pull into your conversation. You never leave this platform; the raw data comes straight through.

Built · Hosted · Managed by Vinkius OpenDataSUS MCP Server - Query Brazilian Health Data

Server ID 019e38cd-f3fa-701f-a8e1-dfa217508524

Vinkius Inspector

Compliance Grade F

Score 3.6/100

Report View Report ↗

Here's how it actually works

The bottom line is you don't navigate the portal; your AI client runs the necessary tools to find, validate, and pull the data for you in a sequence of steps.

Subscribe to OpenDataSUS and (optionally) enter an API Key. Then, use tag_list or organization_list to scope the general area of data you're interested in.

Use package_search to narrow down that scope by topic name, then run package_show on the best match to confirm the dataset's schema and structure.

Finally, feed the required criteria (columns/rows) into datastore_search. The agent executes the query against the official API and returns the raw data payload.

Who is this actually for?

Public health analysts who need current indicators but are tired of manually downloading zip files from departmental websites. Data scientists needing validated datasets for academic research without writing complex API wrappers. Developers building dashboards that require reliable, structured access to official Brazilian government data.

Public Health Analyst

Uses package_search and datastore_search to monitor key health indicators (like vaccination rates) across different regions by running queries against specific resources.

Data Scientist / Researcher

Employs resource_show and tag_list to map the data provenance, understand dataset limitations, and pull raw samples for academic modeling.

Government Developer

Integrates the official health metrics by calling organization_list first, ensuring they only use datasets provided by validated departmental sources before building an application.

What Changes When You Connect

Stop sifting through dozens of departmental pages. Using tag_list lets you scope your research by a general topic (like 'COVID-19') first, then narrow it down with package_search. It’s pure discovery.

Don't guess if the data is clean or formatted right. Run resource_show before querying to see the exact columns and structure of a dataset file, saving you time and headaches.

Skip manual downloads entirely. Once your agent knows which resource package you need, it uses datastore_search to pull filtered tables straight into your chat window for immediate analysis.

Pinpoint exactly who published the data. Use organization_list to ensure the statistics you're citing came from an official source like the Ministério da Saúde.

Validate your whole path with package_show. This tool gives you the full context and metadata for a dataset, confirming its scope and original purpose before you write a single query.

See it in action

01 01

Tracking Vaccine Coverage Changes

A public health analyst needs to see how vaccination rates changed in 2021. They start by running tag_list for 'vacinação' and then use package_search to find the right package. Finally, they tell their agent: 'Query the first 5 rows of that resource using datastore_search,' getting immediate proof of concept data.

02 02

Comparing Data Sources

A developer needs metrics from both the Ministry of Health and a specific state secretariat. They use organization_list to pull all available providers, ensuring they build their application on two distinct, verified data sources rather than just one.

03 03

Verifying Data Columns

A researcher finds a dataset but isn't sure if it has the 'age group' column. Instead of guessing, they use resource_show on the resource ID to verify the schema first. This prevents them from running an empty query and saves hours of debugging.

04 04

Comprehensive Topic Review

A student is writing a paper on general epidemiology. They run package_list, see ten potential datasets, then use group_list to filter those packages by 'Infectious Disease' before selecting the best one for deeper analysis.

The honest tradeoffs

Querying without scoping

Anti-pattern

Just asking: 'Give me COVID data.' The agent fails because it doesn't know which of the thousands of potential datasets you mean. It's too broad.

The Fix

Start by narrowing down. First, run tag_list to find related tags (like 'COVID-19'). Then use those tags in a package_search to get a manageable list of specific packages.

Skipping schema validation

Anti-pattern

Using datastore_search on a resource without first checking its metadata. You might pull data, only to find out the 'date' column is actually text, not a date type.

The Fix

Always run resource_show immediately after finding a promising dataset package. This confirms if the columns you need (like 'data_notificacao') are structured correctly before querying.

Assuming data source authority

Anti-pattern

Pulling data from a resource without knowing which department provided it, leading to unreliable citations or outdated metrics.

The Fix

Always check the origin first. Run organization_list to see all official contributors and confirm that your target dataset belongs to the correct provider.

When It Fits, When It Doesn't

Use OpenDataSUS if your data needs are specific: Brazilian public health statistics, structured by department (Ministério da Saúde) and governed by a CKAN API. It's perfect for researchers who need repeatable queries against official sources.

Don't use this if: 1) You need real-time operational data (e.g., what happened in the last hour at a specific clinic)—this is historical, published data. 2) Your data is private or internal to a single company—this server only accesses public SUS records.

When you are unsure of where to start, don't jump straight to datastore_search. Start with discovery: Run tag_list for conceptual scope, then use organization_list for authority scope. This structured approach guarantees you find the right source before you try to extract data.

Questions you might have

How do I find all possible datasets using package_list? +

You run package_list. This tool provides a comprehensive list of every dataset name available on the portal. It's your starting point for seeing what data exists.

What is the difference between package_search and datastore_search? +

package_search finds the dataset (the container). datastore_search runs the query against the actual rows of data inside that dataset, pulling out the results you need.

Can I check a resource's columns using resource_show? +

Yes. Use resource_show to get technical metadata for any specific file linked to a package. This tells you exactly what columns (like 'municipio') and data types are available.

Does OpenDataSUS cover all Brazilian health data? +

No, it covers the public datasets published on the official OpenDataSUS portal from the Ministry of Health. It won't access private or non-published departmental records.

What happens to my query limits when I use `datastore_search` without an API key? +

You are limited by the default rate caps set by OpenDataSUS. Using an API key bypasses these standard restrictions, letting you run larger, more complex data pulls reliably.

If I need to know who provided a dataset, should I use `organization_list` first? +

Yes, organization_list provides a definitive list of all data providers. You can identify the source organization before running any searches or analyzing specific packages.

What detailed information do I get when I run `package_show`? +

package_show delivers the full metadata package for a dataset. This includes provenance, licensing terms, and structural details—more than just a simple description.

When using `datastore_search`, what are the key parameters I can filter by? +

You can apply filters for date ranges, specific geographic codes, or column values directly in your query. This makes data retrieval highly targeted and efficient.

How can I search for specific rows inside a large CSV dataset? +

You can use the datastore_search tool. Provide the resource_id and use the q parameter for full-text search or the filters parameter to target specific columns.

Can I find which organizations provide the most datasets? +

Yes! Use the organization_list tool to see all data providers registered in the OpenDataSUS portal.

How do I get the download link for a specific data file? +

Use the resource_show tool with the Resource UUID. It will return the metadata including the URL where the file is hosted.

Connect to your AI in seconds.

Datastore search

Group list

Organization list

OpenDataSUS: 8 Tools for Public Health Analytics

Make your AI actually useful.

Datastore Search

Group List

Organization List

Package List

Package Search

Package Show

Resource Show

Tag List

Security and governance baked right in.

Claude AI

Open Claude Settings

Add Custom Connector

Start a conversation

Claude Code

Open your terminal

Add the MCP Server

Start coding

Cursor

One-Click Install (Recommended)

Open Cursor Settings

Add New Server

Use in Composer

Antigravity

Configure Agent Environment

Bind the Endpoint

Execute

VS Code Copilot

One-Click Install (Recommended)

Open MCP Settings

Add Server Config

Windsurf

One-Click Install (Recommended)

Open Windsurf Settings

Add Server Endpoint

LangChain

Install Dependencies

Connect the Server

CrewAI

Define the Tool

Execute Task

Choose How to Get Started

Build Your Own

Make Your AI Do More

Works with Claude, ChatGPT, Cursor, and more

Finding health statistics shouldn't require manual portal hopping.

OpenDataSUS MCP Server: Structured Health Data Discovery

What your AI can actually do with this

Getting Started: Discovering What’s Available

Deep Diving Into Data Structure

Querying the Raw Data

The Workflow Summary

Here's how it actually works

Who is this actually for?

What Changes When You Connect

See it in action

Tracking Vaccine Coverage Changes

Comparing Data Sources

Verifying Data Columns

Comprehensive Topic Review

The honest tradeoffs

Querying without scoping

Skipping schema validation

Assuming data source authority

When It Fits, When It Doesn't

Questions you might have