How to Use the CTA MCP in Pydantic AI

Q: How does Pydantic AI handle a CTA bus that's delayed and not sending real-time data?

When your Pydantic AI agent calls getbuspredictions, the response includes a boolean field indicating if the bus is real-time tracked. If it's just a scheduled prediction, your agent knows not to present it as a live, up-to-the-second ETA.

Q: Can my Pydantic AI agent get all CTA alerts, or just for one route?

Both. The getservicealerts tool is flexible. Your Pydantic AI agent can call it without parameters to get all system-wide alerts, or provide a route ID to get alerts only for that specific bus or train line.

Get type-safe, validated CTA transit data for your Pydantic AI agent and build applications that don't break.

See Vinkius in Action

Works with every AI agent you already use

…and any MCP-compatible client

MCP Servers - Free for Subscribers

Connect CTA MCP to Pydantic AI

Create your Vinkius account to connect CTA to Pydantic AI 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 CTA with Pydantic AI

Ask AI about this MCP

ChatGPT

Claude

Perplexity

Get Data You Can Actually Trust

This isn't your typical API. When your agent calls `get_bus_predictions` or `get_train_arrivals`, the response is automatically validated against a Pydantic model. If the CTA API ever sends back a malformed field or an unexpected null value, your agent will raise a `ValidationError` immediately. This means no silent failures or corrupted data downstream. Your Pydantic AI agent operates with certainty. You know the vehicle ID you get from `get_bus_vehicles` is a string and the latitude is a float, because the MCP toolset guarantees it.

A Type-Safe MCP Server for Pydantic AI

Build complex trip-planning logic without guesswork. The tools are designed to work together. Your agent can use `get_bus_routes` to get a list of valid `routeId`s, then pass one to `get_bus_stops` to get a list of `stpid`s. Each step is clear and the data types are enforced. This structured approach prevents common errors. For instance, an agent can't accidentally ask for predictions for a stop that doesn't exist on a given route. By using the suite of tools (`get_route_directions`, `get_stop_details`), your Pydantic AI agent can build a valid and correct model of the CTA network before it even checks for live arrivals.

Handle Service Issues Without Crashing

Real-world transit has problems. The `get_service_alerts` tool delivers structured data on delays, reroutes, and disruptions. Your agent gets the severity, cause, and a description, all neatly parsed into a Pydantic model. Instead of trying to parse a messy string, your agent can programmatically check the `status` field from `get_route_status`. If it sees 'SERVICE DISRUPTION', it knows to then call `get_service_alerts` for details. This lets you build a robust agent that informs users about problems instead of just failing.

Setup guide

Set up CTA MCP in Pydantic AI

Prerequisites

Python 3.10+ installed
pydantic-ai-slim[fastmcp] package
Active Vinkius subscription with a valid endpoint token

1

Install Pydantic AI with FastMCP
Run pip install "pydantic-ai-slim[fastmcp]". The FastMCP toolset replaces the deprecated MCPServerHTTP class with full protocol support.
2

Configure the FastMCPToolset
Pass a JSON-style config dict to FastMCPToolset with your Vinkius URL. Replace [YOUR_TOKEN_HERE] with your token from cloud.vinkius.com. Supports Streamable HTTP, SSE, and Stdio transports.
3

Create and run your agent
Pass the toolset to Agent(toolsets=[toolset]) and call agent.run(). Swap openai:gpt-4o for any supported model — Anthropic, Google, Mistral, or Groq.

agent.py

from pydantic_ai import Agent
from pydantic_ai.toolsets.fastmcp import FastMCPToolset

toolset = FastMCPToolset({
    "mcpServers": {
        "cta-mcp": {
            "url": "https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp"
        }
    }
})

agent = Agent(
    "openai:gpt-4o",
    toolsets=[toolset],
    system_prompt="You have access to CTA tools.",
)

result = await agent.run("List recent CTA transactions")
print(result.output)

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 CTA. 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 CTA 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 CTA MCP in Pydantic AI

When your Pydantic AI agent calls `get_bus_predictions`, the response includes a boolean field indicating if the bus is real-time tracked. If it's just a scheduled prediction, your agent knows not to present it as a live, up-to-the-second ETA.

Both. The `get_service_alerts` tool is flexible. Your Pydantic AI agent can call it without parameters to get all system-wide alerts, or provide a route ID to get alerts only for that specific bus or train line.

Not silently. Because you're using this MCP Server with Pydantic AI, any unexpected change in the CTA's data structure will cause a `ValidationError`. Your agent will fail loudly, telling you exactly which field was wrong, so you can fix it instead of processing bad data.

That's the one piece of information this MCP server doesn't provide directly. The MapIds are part of the CTA's static GTFS data feed, which you'd typically download separately and store. For common stations, the documentation provides examples.

The agent processes public CTA data: stop locations, route geometries, and real-time vehicle positions. Security is handled by Vinkius's zero-trust architecture. Your connection is authenticated with a single token, and the server itself has no access to your broader systems or data.

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