MDIC (Comércio Exterior) MCP for AI. Query official Brazilian trade statistics instantly.

Q: How do I find out what datasets are available using searchpackages?

Run searchpackages and pass the general topic, like 'exportação'. This tool filters through all MDIC packages to give you a shortlist of relevant dataset names.

Q: Do I need getpackage before searchingdatastore?

No. While getpackage gives detailed metadata, searchdatastore is the action tool. You typically use discovery tools first to find the package ID, and then pass that ID directly into searchdatastore.

Q: What does getresource do?

getresource pulls metadata for a specific data file within a dataset. Use it if you need to confirm the exact download link or format for a resource ID you found earlier.

Q: Can I query any date range using searchdatastore?

Yes, searchdatastore accepts SQL-like parameters. You specify your criteria (e.g., 'Year > 2020 AND Month = 1') to limit the rows returned.

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.

MDIC (Comércio Exterior) connects your AI client directly to Brazil's Ministry of Development, Industry, Trade and Services (MDIC) open data portal.

It lets you list all available trade datasets, search for specific packages by keyword, inspect dataset metadata, and query the full datastore using SQL-like parameters for export/import statistics.

What your AI can do

Get package

Gets detailed metadata for a single, specified data package (dataset).

Get resource

Retrieves metadata and links for a specific file resource within an MDIC package.

List packages

Lists every single available dataset (package) in the entire MDIC portal.

+ 2 more capabilities included

List all available datasets

Retrieves a full inventory of every trade data package hosted on the MDIC portal.

Search for specific dataset packages

Filters and finds relevant data packages using keywords like 'exportação' or 'importação'.

Get metadata for a package

Retrieves detailed information, tags, and resource lists for one known dataset ID.

Get file details and links

Provides metadata and download paths for individual data files within a selected package.

Run filtered queries on data rows

Executes SQL-like searches directly against the datastore to pull specific, limited rows of trade statistics.

Ask an AI about this

Included with Plan

Waiting for input…

AI Agent

MDIC (Comércio Exterior) MCP Server: 5 Tools for Data Retrieval

These five tools let your AI client discover, inspect, and run queries against the massive official Brazilian trade data repository.

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 MDIC (Comércio Exterior) on Vinkius

Get Package

Gets detailed metadata for a single, specified data package (dataset).

Get Resource

Retrieves metadata and links for a specific file resource within an MDIC package.

List Packages

Lists every single available dataset (package) in the entire MDIC portal.

Search Datastore

Runs a query directly against the data rows to pull filtered results from a resource.

Search Packages

Finds relevant dataset packages using a keyword search string.

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 MDIC (Comércio Exterior) 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": {
  "mdic-comercio-exterior": {
    "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 MDIC (Comércio Exterior) 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:

"mdic-comercio-exterior": {
  "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 MDIC (Comércio Exterior), 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

MDIC (Comércio Exterior) MCP server cover

Independent Platform Disclaimer: Vinkius is an independent platform and is not affiliated with, endorsed by, sponsored by, verified by, or otherwise authorized by MDIC (Comércio Exterior). 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 5 powerful capabilities that interface natively with Claude, ChatGPT, Cursor, and other compatible AI platforms. No middleware. No custom integration required.

Sifting through government trade portals is a nightmare.

Before this server, analyzing Brazilian foreign trade meant navigating sprawling web portals. You’d spend hours clicking between different packages, manually downloading massive CSV files for exports and imports separately, then opening Excel just to consolidate the dates and figures you needed. It was copy-paste hell.

Now, your agent handles it all. You tell it what data point you need—say, 'Export value from Shanghai in Q3'—and it uses `search_packages` to find the right dataset, then runs a precise query with `search_datastore`. The result is structured text, not 50 megabytes of mess.

MDIC (Comércio Exterior) MCP Server: Query Official Brazilian Trade Data

The painful manual steps that disappear are the web browsing, the file download process, and the subsequent data cleaning. No more guessing which CSV column means what; the server handles all the resource mapping.

What's different now is control. You get direct, programmatic access to the raw data structure, allowing you to build automated reports in code rather than spending days wrestling with spreadsheets.

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

WHO IT'S FOR: This server connects your AI client straight into Brazil’s Ministry of Development, Industry, Trade and Services (MDIC) open data portal. You use it when you need official Brazilian foreign trade statistics—you wanna pull the numbers directly from the source without downloading a mountain of files.

How It Works: Your agent handles the entire process: finding the right dataset, checking its scope, and running complex queries against the actual data rows. You never have to deal with manual CSV downloads again.

Finding Datasets (Discovery): First, you need to know what's available. Use list_packages if you want a complete inventory of every single trade dataset package hosted on the MDIC portal. If that list is too long and you know what keywords you're after—like 'exportação' or 'importação'—you can use search_packages. That narrows down the results instantly, pointing your agent right to relevant packages.
Inspecting Data (Metadata): Once you've identified a package ID, don't just assume it has what you need. You run get_package to pull detailed metadata for that specific dataset. This gives you tags and confirms the full scope of the data before you commit to running a query. If you then want to check the individual files within that confirmed package, get_resource provides all the necessary metadata and download paths for those resources.
Querying the Data (Extraction): This is where the money's at. You don’t just pull whole tables; you pull specific metrics. Using search_datastore, your agent executes queries that function like SQL directly against the underlying data rows. You can filter and limit the results to pull only the exact trade statistics—like specific export volumes or import values—that fit your criteria, making sure you get clean, usable numbers straight into your workflow.

Built · Hosted · Managed by Vinkius MDIC MCP Server - Analyze Brazilian Trade Data

Server ID 019e38bf-d9d9-72dc-8843-6b17bde488f3

Vinkius Inspector

Compliance Grade A+

Score 100/100

Report View Report ↗

Who is this actually for?

This server is for data professionals who regularly work with large, official government datasets. If you're an economist or a BI lead tired of manually downloading massive CSVs from web portals and spending hours cleaning up inconsistent files, this is for you. You need to analyze trends in Brazilian trade, not just look at static dashboards.

Data Analyst

Uses search_datastore to run precise queries against specific date ranges or product codes within the raw data.

Economist

Employs natural language prompts to analyze official trade volumes, using the full dataset discovery capability of list_packages.

Business Intelligence Lead

Integrates structured government statistics into automated reports by calling out specific resource IDs via get_resource.

What Changes When You Connect

Find the right data faster. Instead of clicking through dozens of links, use search_packages to filter datasets by terms like 'exportação' or 'comex-stat'.

Skip manual downloads entirely. You don't grab a CSV; you ask your agent to run a query using search_datastore, and the structured data comes back instantly.

Full transparency on resources. Use get_resource when you need confirmation of specific file IDs or download links before integrating them into a script.

Discover every dataset available. If you don't know what data exists, start with list_packages to get the full inventory of all trade packages.

Analyze trends, not just records. The server handles complex joins and filtering via search_datastore, letting your agent focus on summarizing results.

See it in action

01 01

Tracking a specific product's export history

A user needs to see how the value of Product Code X changed over three years. They ask their agent, and the agent runs search_packages for 'comex-stat', then uses that ID with search_datastore to filter by the specific product code and date range, delivering only the needed rows.

02 02

Identifying all available trade data

A new analyst joins the team and needs a complete list of everything MDIC offers. They simply trigger list_packages, which immediately provides every dataset name, eliminating manual portal browsing.

03 03

Verifying file availability

An engineer knows they need the 'Monthly Exports' data but can't find the exact download link in the documentation. They run get_package first to confirm the dataset, then use that ID with get_resource to get the precise metadata and working URL.

04 04

Comparing multiple market sectors

A researcher wants to compare trade data across three different geographic regions. They ask their agent to run targeted queries using search_datastore, passing in the necessary filters for all three areas simultaneously, getting a single comparative result set.

The honest tradeoffs

Searching by memory

Anti-pattern

Trying to remember if you need to search the dataset list or search the data itself. You might try to pass keywords directly into search_datastore and fail.

The Fix

First, use search_packages to find the correct dataset name (package). Then, use that package ID with search_datastore to run your query on the actual numbers.

Using metadata for data

Anti-pattern

Assuming that just knowing a resource exists (get_resource) is enough. You get the file link but still can't analyze the content without running a query.

The Fix

After finding a resource, always confirm you need to run actual numbers against it. Use search_datastore to execute the filtering and retrieval action.

Over-listing resources

Anti-pattern

When only needing one dataset, running list_packages just to see everything. This is slow and creates unnecessary noise.

The Fix

If you know what you're looking for (e.g., 'comex'), use search_packages instead of listing every single package available.

When It Fits, When It Doesn't

Use this server if your workflow requires highly structured, filtered data retrieval from official government sources. You need to know what the numbers are (e.g., 'Value FOB for product X in month Y').

Don't use it if you just want general context or a summary of trade policy—your AI client should handle that narrative part. If your goal is simply to list all datasets, start with list_packages. But if the end result must be a filtered table of numbers, you will always need to follow discovery (search_packages) -> inspection (get_package) -> execution (search_datastore). This structure guarantees maximum control over the data path.

Questions you might have

How do I find out what datasets are available using search_packages? +

Run search_packages and pass the general topic, like 'exportação'. This tool filters through all MDIC packages to give you a shortlist of relevant dataset names.

Do I need get_package before searching_datastore? +

No. While get_package gives detailed metadata, search_datastore is the action tool. You typically use discovery tools first to find the package ID, and then pass that ID directly into search_datastore.

What does get_resource do? +

get_resource pulls metadata for a specific data file within a dataset. Use it if you need to confirm the exact download link or format for a resource ID you found earlier.

Can I query any date range using search_datastore? +

Yes, search_datastore accepts SQL-like parameters. You specify your criteria (e.g., 'Year > 2020 AND Month = 1') to limit the rows returned.

When using `list_packages` or other tools repeatedly, how do I handle API rate limits? +

You must provide an API key for high-volume usage. The server supports keys to bypass standard rate limits imposed by the MDIC portal. If you don't include a key, your AI client will receive a 429 error response when hitting predefined request thresholds.

If I need to filter data from multiple criteria (e.g., Year AND Product Code) using `search_datastore`, how should I structure the query? +

You use standard SQL-like syntax for filtering within the tool's parameters. You combine conditions using logical operators like 'AND' or 'OR'. For example, specifying WHERE year = 2023 AND product_code = 'XYZ'.

What specific metadata does `get_package` return about a dataset? +

It returns core information including the package title, description, list of contained resources (data files), and associated tags. This allows your AI client to understand the scope and purpose of the entire dataset before querying any single resource.

What happens if I use `search_packages` with an invalid or misspelled query string? +

The tool returns a structured error message detailing the validation failure. It won't crash; instead, it tells your agent exactly which part of the search input was malformed and why, allowing for immediate correction.

Can I search for specific trade terms like 'soybean' or 'iron ore'? +

Yes! Use the search_packages tool with your query string. It will return all datasets matching those terms within the MDIC portal.

How do I access the actual data rows inside a CSV resource? +

Use the search_datastore tool by providing the resource_id. You can also apply filters and limits to retrieve exactly the data points you need.

Is it possible to see the file format and download URL for a dataset? +

Yes. The get_package tool returns a list of resources, and get_resource provides specific metadata including the format (CSV, XLSX, etc.) and the direct download URL.

Connect to your AI in seconds.

Get package

Get resource

List packages

MDIC (Comércio Exterior) MCP Server: 5 Tools for Data Retrieval

Make your AI actually useful.

Get Package

Get Resource

List Packages

Search Datastore

Search Packages

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

Sifting through government trade portals is a nightmare.

MDIC (Comércio Exterior) MCP Server: Query Official Brazilian Trade Data

What your AI can actually do with this

Here's how it actually works

Who is this actually for?

What Changes When You Connect

See it in action

Tracking a specific product's export history

Identifying all available trade data

Verifying file availability

Comparing multiple market sectors

The honest tradeoffs

Searching by memory

Using metadata for data

Over-listing resources

When It Fits, When It Doesn't

Questions you might have