CARTO MCP Server for OpenAI Agents SDK 10 tools — connect in under 2 minutes
The OpenAI Agents SDK enables production-grade agent workflows in Python. Connect CARTO through Vinkius and your agents gain typed, auto-discovered tools with built-in guardrails. no manual schema definitions required.
ASK AI ABOUT THIS MCP SERVER
Vinkius supports streamable HTTP and SSE.
import asyncio
from agents import Agent, Runner
from agents.mcp import MCPServerStreamableHttp
async def main():
# Your Vinkius token. get it at cloud.vinkius.com
async with MCPServerStreamableHttp(
url="https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp"
) as mcp_server:
agent = Agent(
name="CARTO Assistant",
instructions=(
"You help users interact with CARTO. "
"You have access to 10 tools."
),
mcp_servers=[mcp_server],
)
result = await Runner.run(
agent, "List all available tools from CARTO"
)
print(result.final_output)
asyncio.run(main())
* Every MCP server runs on Vinkius-managed infrastructure inside AWS - a purpose-built runtime with per-request V8 isolates, Ed25519 signed audit chains, and sub-40ms cold starts optimized for native MCP execution. See our infrastructure
About CARTO MCP Server
Connect your CARTO platform to any AI agent and take full control of your cloud-native spatial analytics without touching the GIS interface.
The OpenAI Agents SDK auto-discovers all 10 tools from CARTO through native MCP integration. Build agents with built-in guardrails, tracing, and handoff patterns. chain multiple agents where one queries CARTO, another analyzes results, and a third generates reports, all orchestrated through Vinkius.
What you can do
- Spatial SQL & Jobs — Command explicit SQL queries running directly against your BigQuery, Snowflake, or Redshift warehouse. Execute quick reads or spawn long-running, asynchronous batch transformations.
- Geocoding & Batching — Convert unstructured strings into precise lat/lon points. Bulk-process 100s of addresses using CARTO's native Location Data Services (LDS) gracefully.
- Isolines & Reachability — Calculate travel-time or distance polygons around origin points to instantly graph accessible zones via cars or walking.
- Routing — Generate optimal vector geometry paths connecting two points on the globe, measuring both physical distance and travel time securely.
- Data Management — Instruct your agent to list active mapping datasets or import massive external CSV/GeoJSON files directly via public URLs.
The CARTO MCP Server exposes 10 tools through the Vinkius. Connect it to OpenAI Agents SDK in under two minutes — no API keys to rotate, no infrastructure to provision, no vendor lock-in. Your configuration, your data, your control.
How to Connect CARTO to OpenAI Agents SDK via MCP
Follow these steps to integrate the CARTO MCP Server with OpenAI Agents SDK.
Install the SDK
Run pip install openai-agents in your Python environment
Replace the token
Replace [YOUR_TOKEN_HERE] with your Vinkius token from cloud.vinkius.com
Run the script
Save the code above and run it: python agent.py
Explore tools
The agent will automatically discover 10 tools from CARTO
Why Use OpenAI Agents SDK with the CARTO MCP Server
OpenAI Agents SDK provides unique advantages when paired with CARTO through the Model Context Protocol.
Native MCP integration via `MCPServerSse`, pass the URL and the SDK auto-discovers all tools with full type safety
Built-in guardrails, tracing, and handoff patterns let you build production-grade agents without reinventing safety infrastructure
Lightweight and composable: chain multiple agents and MCP servers in a single pipeline with minimal boilerplate
First-party OpenAI support ensures optimal compatibility with GPT models for tool calling and structured output
CARTO + OpenAI Agents SDK Use Cases
Practical scenarios where OpenAI Agents SDK combined with the CARTO MCP Server delivers measurable value.
Automated workflows: build agents that query CARTO, process the data, and trigger follow-up actions autonomously
Multi-agent orchestration: create specialist agents. one queries CARTO, another analyzes results, a third generates reports
Data enrichment pipelines: stream data through CARTO tools and transform it with OpenAI models in a single async loop
Customer support bots: agents query CARTO to resolve tickets, look up records, and update statuses without human intervention
CARTO MCP Tools for OpenAI Agents SDK (10)
These 10 tools become available when you connect CARTO to OpenAI Agents SDK via MCP:
calculate_isoline
The range parameter is in seconds for time-based isolines. Returns a GeoJSON polygon representing the reachable area. Use for service area analysis, store catchment zones, and logistics planning. Generate travel-time or travel-distance isoline polygons from a center point using the CARTO LDS Isolines API, producing reachability contours showing areas accessible within a specified time or distance threshold
calculate_route
Returns the route as GeoJSON with total distance (meters) and duration (seconds). Consumes LDS routing credits. Calculate the optimal driving route between two points using the CARTO LDS Routing API, returning distance, duration, and route geometry suitable for visualization on CARTO maps
create_async_sql_job
"}`. Returns a job_id that can be polled for completion status. Use for ETL operations, materialized view refreshes, and heavy geospatial computations. The job runs in your data warehouse and results are stored there. Submit a long-running SQL query as an asynchronous batch job via the CARTO SQL Job API, suitable for heavy spatial analytics, large table transformations, and complex multi-join operations that exceed the 60-second synchronous timeout
execute_sql_query
carto.com/api/v2/sql?q=`. The query runs synchronously with a 1-minute timeout. Use for quick analytical queries, spatial joins, and data exploration. For long-running queries exceeding 60 seconds, use the async job endpoint instead. Supports PostGIS/BigQuery spatial functions natively. Execute an arbitrary SQL query against your CARTO data warehouse connection using the SQL API v2, returning results as JSON rows directly from BigQuery, Snowflake, Redshift, or PostgreSQL
geocode_address
Returns latitude, longitude, and formatted address. Consumes LDS geocoding credits from your CARTO plan. Use sparingly for individual lookups; for bulk operations use the batch endpoint instead. Forward-geocode a single address string into geographic coordinates using the CARTO Location Data Services (LDS) geocoding endpoint, powered by TomTom or HERE depending on your CARTO plan configuration
geocode_batch_addresses
Designed for bulk processing of customer lists, store locators, and CRM datasets. Consumes LDS credits per address. Returns an array of geocoded results with match quality indicators. Batch-geocode multiple addresses in a single request using the CARTO LDS batch geocoding API, efficiently converting large address lists into coordinates without making individual API calls per address
get_import_status
Poll periodically until state becomes "complete" or "failure". On success, the response includes the table_name of the newly created dataset in your warehouse. Check the status of a previously initiated CARTO data import job, returning progress percentage, current state (uploading, importing, complete, failure), and any error details if the import encountered issues
import_external_file
"}`. Supports CSV, GeoJSON, Shapefile (zipped), KML, GPX, and Excel files. Returns an import_id for status tracking. The file is downloaded, parsed, and loaded into your connected data warehouse. Import an external data file (CSV, GeoJSON, Shapefile, KML) into your CARTO data warehouse by providing a publicly accessible URL, creating a new managed table that can be used for spatial analysis and visualization
list_map_datasets
Returns an array of dataset/visualization objects. Use to discover available data layers, check dataset freshness, and audit organization assets. List all visualization datasets (maps and tables) available in your CARTO organization, returning metadata including creation dates, privacy settings, table names, and row counts
poll_async_job_status
Poll periodically (every 5-10 seconds) until status changes to "done" or "failed". The response includes created_at, updated_at, and the original query for audit purposes. Check the execution status of a previously submitted CARTO async SQL job, returning the current state (pending, running, done, failed) and any error messages if the job encountered issues
Example Prompts for CARTO in OpenAI Agents SDK
Ready-to-use prompts you can give your OpenAI Agents SDK agent to start working with CARTO immediately.
"Execute a SQL query limiting to 10 rows on my 'retail_stores' dataset to check the schema."
"Take these 5 addresses in Madrid and bulk geocode them to lat/lon coordinates."
"Generate a 15-minute drive-time isoline around Times Square, New York."
Troubleshooting CARTO MCP Server with OpenAI Agents SDK
Common issues when connecting CARTO to OpenAI Agents SDK through the Vinkius, and how to resolve them.
MCPServerStreamableHttp not found
pip install --upgrade openai-agentsAgent not calling tools
CARTO + OpenAI Agents SDK FAQ
Common questions about integrating CARTO MCP Server with OpenAI Agents SDK.
How does the OpenAI Agents SDK connect to MCP?
MCPServerSse(url=...) to create a server connection. The SDK auto-discovers all tools and makes them available to your agent with full type information.Can I use multiple MCP servers in one agent?
MCPServerSse instances to the agent constructor. The agent can use tools from all connected servers within a single run.Does the SDK support streaming responses?
Connect CARTO with your favorite client
Step-by-step setup guides for every MCP-compatible client and framework:
Anthropic's native desktop app for Claude with built-in MCP support.
AI-first code editor with integrated LLM-powered coding assistance.
GitHub Copilot in VS Code with Agent mode and MCP support.
Purpose-built IDE for agentic AI coding workflows.
Autonomous AI coding agent that runs inside VS Code.
Anthropic's agentic CLI for terminal-first development.
Python SDK for building production-grade OpenAI agent workflows.
Google's framework for building production AI agents.
Type-safe agent development for Python with first-class MCP support.
TypeScript toolkit for building AI-powered web applications.
TypeScript-native agent framework for modern web stacks.
Python framework for orchestrating collaborative AI agent crews.
Leading Python framework for composable LLM applications.
Data-aware AI agent framework for structured and unstructured sources.
Microsoft's framework for multi-agent collaborative conversations.
Connect CARTO to OpenAI Agents SDK
Get your token, paste the configuration, and start using 10 tools in under 2 minutes. No API key management needed.