How to Use the New York Times MCP in Pydantic AI

Q: How does Pydantic AI guarantee the structure of New York Times API responses?

When your agent uses a tool like searcharticles, Pydantic AI intercepts the JSON response from the MCP server. It then parses this data into a Pydantic model, and if any field is missing or has the wrong type, it raises an immediate error.

Q: What is the setup process for Pydantic AI and the New York Times MCP server?

It's straightforward. After pip install "pydantic-ai-slimmcp", you create an MCPToolset instance with your Vinkius server URL. Then you pass this toolset into the toolsets list when you initialize your Agent.

Get type-safe, validated New York Times data from this MCP Server in any Python agent with Pydantic AI. Stop guessing what the API will return.

See Vinkius in Action

Works with every AI agent you already use

…and any MCP-compatible client

MCP Servers - Free for Subscribers

Connect New York Times MCP to Pydantic AI

Create your Vinkius account to connect New York Times 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 New York Times with Pydantic AI

Ask AI about this MCP

ChatGPT

Claude

Perplexity

Search with Guaranteed Structure

The `search_articles` tool is your agent's entry point for finding news. It takes a query, optional dates, and a sort order, then returns a list of articles. With Pydantic AI, that returned list isn't a guess—it's a promise. Each article object is validated against a Pydantic model at runtime. If the NYT API ever changes a field or sends bad data, your code will raise a `ValidationError` immediately instead of failing silently later.

Model-Agnostic Cultural Data

Give your agent access to cultural trends with `get_book_lists` and `get_movie_reviews`. Your agent can ask for the 'hardcover-fiction' list or search for reviews of a specific film. Because Pydantic AI is model-agnostic, you can use these tools with OpenAI, Gemini, or a local model. The framework handles the tool-calling logic, and you get validated Pydantic objects back, no matter which LLM is driving the agent.

A Correctness-First MCP Server

This server exposes tools for checking popular content, like `get_most_emailed` and `get_most_shared`. They are simple to call, taking just a time period of 1, 7, or 30 days. The whole point of Pydantic AI is correctness. By passing the `MCPToolset` to your agent, you ensure every response from these tools is parsed and validated. You're not just getting data; you're getting data you can trust.

Setup guide

Set up New York Times 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": {
        "new-york-times-mcp": {
            "url": "https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp"
        }
    }
})

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

result = await agent.run("List recent New York Times 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 New York Times. 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 New York Times 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 New York Times MCP in Pydantic AI

When your agent uses a tool like `search_articles`, Pydantic AI intercepts the JSON response from the MCP server. It then parses this data into a Pydantic model, and if any field is missing or has the wrong type, it raises an immediate error.

Yes. Pydantic AI is model-agnostic. You configure your agent to use your local model, then provide the New York Times `MCPToolset`. The agent will call `get_top_stories` and you'll get a validated Pydantic object back.

It's straightforward. After `pip install "pydantic-ai-slim[mcp]"`, you create an `MCPToolset` instance with your Vinkius server URL. Then you pass this toolset into the `toolsets` list when you initialize your `Agent`.

This method handles the tool definition, prompting, and response validation for you. Instead of writing boilerplate to call an endpoint and parse JSON, you just give the tool to an agent and get a clean, type-hinted Python object back.

The server only handles the movie review content and metadata provided by the NYT API. Pydantic AI validates this data on your client, but the server itself is ephemeral. It doesn't log or retain your search queries or the review text it processes.

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