How to Use the CARTO MCP in OpenAI Agents SDK

Q: Can my OpenAI Agents SDK system run synchronous CARTO queries?

Yes, the executesqlquery tool handles quick analytical pulls under 60 seconds. The agent receives JSON rows straight from your PostgreSQL or Redshift instance. Anything longer needs the async batch endpoints.

Q: What happens if an agent fails a CARTO import job?

The getimportstatus tool returns a failure state along with the specific error string. You configure your OpenAI agent to catch that failure and retry the upload. Tracing captures the exact moment the file parsing broke.

Build production-grade spatial agents with OpenAI Agents SDK that run heavy SQL and routing queries directly against your data warehouse.

See Vinkius in Action

Works with every AI agent you already use

…and any MCP-compatible client

MCP Servers - Free for Subscribers

Connect CARTO MCP to OpenAI Agents SDK

Create your Vinkius account to connect CARTO to OpenAI Agents SDK and route execution through our secure gateway. The platform manages server hosting, runtime updates, and security layers. Configuration requires no manual server provisioning.

GDPR Free for Subscribers

Setup CARTO with OpenAI Agents SDK

Ask AI about this MCP

ChatGPT

Claude

Perplexity

Spatial analysis within OpenAI Agents SDK

Your agent triggers `calculate_isoline` to map out 600-second drive-time polygons around retail locations. It passes those coordinates directly into a guardrailed handoff where another agent validates the bounds. You get exact GeoJSON outputs fed straight into your OpenAI tracing dashboard. Managing point-to-point logistics works exactly the same way. The agent fires `calculate_route` to pull driving distances in meters and durations in seconds. Because this MCP Server runs on Vinkius, your OpenAI agent handles the spatial math without exposing your raw coordinates to an unmanaged environment.

Asynchronous warehouse queries

Calling `create_async_sql_job` pushes heavy spatial joins directly to your connected data warehouse instead of timing out the agent. OpenAI's execution loop simply pauses while the warehouse crunches the multi-join operation. You avoid blocking the main thread during massive ETL tasks. Checking progress takes a quick `poll_async_job_status` call every ten seconds. Once the status flips to done, your agent pulls the aggregated results for the next step in its pipeline. This setup keeps your production system stable even when processing millions of geographic rows.

Automated pipeline ingestion

Pointing `import_external_file` at a zipped Shapefile URL forces CARTO to download, parse, and load the geometry into a new managed table. Your ingestion agent handles the entire process autonomously. It just monitors `get_import_status` until the upload hits 100 percent. Discovering what already exists requires a quick `list_map_datasets` request. The agent reads the creation dates and row counts to decide if it needs to fetch fresh data. You let the system manage its own spatial inventory while OpenAI logs every decision.

Setup guide

Set up CARTO MCP in OpenAI Agents SDK

Prerequisites

Python 3.10+ installed
openai-agents package (pip install openai-agents)
Active Vinkius subscription with a valid endpoint token

1

Install the SDK

Run pip install openai-agents to install the OpenAI Agents SDK. The MCP integration is built-in — no extra dependencies needed.
2

Connect via SSE transport

Use MCPServerSse with your Vinkius endpoint URL. Replace [YOUR_TOKEN_HERE] with your token from cloud.vinkius.com. The SDK auto-discovers all CARTO tools at runtime.
3

Create your Agent

Pass the MCP to Agent(mcp_servers=[server]). The agent receives CARTO tools as native definitions — JSON schemas resolve automatically.
4

Run the agent

Call Runner.run(agent, prompt) to execute. The agent invokes the appropriate CARTO tools and returns structured results. Copy the full example on the right to get started.

agent.py

import asyncio
from agents import Agent, Runner
from agents.mcp import MCPServerSse

async def main():
    async with MCPServerSse(
        url="https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp"
    ) as server:
        agent = Agent(
            name="CARTO Agent",
            instructions="You have access to CARTO tools.",
            mcp_servers=[server],
        )
        result = await Runner.run(agent, "List recent transactions")
        print(result.final_output)

asyncio.run(main())

Get your connection token →

Independent Platform Disclaimer: Vinkius is an independent platform and is not affiliated with, endorsed by, sponsored by, verified by, or otherwise authorized by CARTO. 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.

Why Choose Vinkius

Vinkius connects your tools to AI with real-time monitoring and automatic cost savings — all from one dashboard.

Connect CARTO now

Real-time monitoring

Live

visibility into every interaction

Connect your favorite tools to your AI and see exactly what's happening — every request, every response, in real time.

Built-in savings

60%

lower AI costs

Vinkius compresses data between your apps and your AI automatically. Lower bills every month — no configuration required.

Single dashboard

One

place for every integration

Every tool your AI connects to, managed from a single screen. One account, complete control.

Common questions about CARTO MCP in OpenAI Agents SDK

Install the openai-agents package first. You initialize the MCP connection using MCPServerStreamableHttp with your Vinkius endpoint URL and pass it to the Agent constructor. Setting cacheToolsList=True drops the initialization latency on your end.

Yes, the execute_sql_query tool handles quick analytical pulls under 60 seconds. The agent receives JSON rows straight from your PostgreSQL or Redshift instance. Anything longer needs the async batch endpoints.

Your agent feeds an array of customer addresses into geocode_batch_addresses to process the entire list at once. This consumes LDS credits but avoids spamming the single-address endpoint. The response includes match quality indicators for every row.

The get_import_status tool returns a failure state along with the specific error string. You configure your OpenAI agent to catch that failure and retry the upload. Tracing captures the exact moment the file parsing broke.

Your proprietary shapefiles and customer coordinate data never touch the client directly during processing. Vinkius runs the connection inside a V8 Isolate Sandbox that spins down the moment the query finishes. You issue one endpoint token and zero-trust architecture handles the rest.

Use it with your favorite AI tools

Connect this server to Cursor, Claude, VS Code, and more.

OpenAI Agents SDK sdk-python

Google ADK sdk-python

Pydantic AI sdk-python

Vercel AI SDK sdk-typescript