How to Use the Harvard ClinicalTrials MCP in Pydantic AI

Q: How do I register the Harvard ClinicalTrials MCP Server in Pydantic AI?

Use the MCPToolset class with your server URL. Pass it to the toolsets parameter when initializing your Agent, and it will automatically map all 16 endpoints.

Q: What happens if a Harvard ClinicalTrials query fails in Pydantic AI?

The framework throws a validation error immediately. If searchbylocation returns malformed JSON or unexpected fields, the agent stops execution rather than hallucinating an answer.

Q: How do I find active patient recruitment data?

Your agent calls the searchrecruiting tool. It returns only the trials currently accepting patients, which you can then filter further by condition or sponsor.

Type-safe clinical trial discovery for Pydantic AI. Force Harvard ClinicalTrials data into strict Python models and catch bad API responses instantly.

See Vinkius in Action

Works with every AI agent you already use

…and any MCP-compatible client

MCP Servers - Free for Subscribers

Connect Harvard ClinicalTrials MCP to Pydantic AI

Create your Vinkius account to connect Harvard ClinicalTrials 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 Harvard ClinicalTrials with Pydantic AI

Ask AI about this MCP

ChatGPT

Claude

Perplexity

Validate public trial data at runtime

The Harvard ClinicalTrials MCP Server exposes 16 government database endpoints directly to your Python application. When you connect this to Pydantic AI, tools like `get_study` and `search_by_condition` return data that gets validated against your exact schemas at runtime. If the public registry changes a field name or returns an unexpected string, your agent fails loudly. You never have to worry about silent data corruption when pulling records via `search_fda_regulated` or `search_pediatric`. The framework guarantees the types match before your code proceeds.

Map study timelines to strict models

Extracting trial timelines is handled by the `get_study_timeline` tool. You can define a Pydantic model for milestone dates, and the agent will map the raw JSON into your structured format. Because Pydantic AI is model-agnostic, you can swap between Anthropic for complex protocol analysis and a local model for basic `search_by_sponsor` lookups. The MCP toolset behaves exactly the same regardless of which LLM is driving the logic.

Target specific medical interventions

Finding specific medical treatments requires the `search_by_intervention` and `search_device_trials` tools. The agent queries the exact drug or implant name and returns the matching NCT IDs. Setup uses the unified `MCPToolset` class. You point it at your server URL, pass it to your Agent's `toolsets` array, and the system handles the rest. You get production-grade reliability when parsing complex trial results with `get_study_results`.

Setup guide

Set up Harvard ClinicalTrials 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": {
        "harvard-clinicaltrials-mcp": {
            "url": "https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp"
        }
    }
})

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

result = await agent.run("List recent Harvard ClinicalTrials 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 ClinicalTrials.gov. 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 Harvard ClinicalTrials 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 Harvard ClinicalTrials MCP in Pydantic AI

Use the `MCPToolset` class with your server URL. Pass it to the `toolsets` parameter when initializing your Agent, and it will automatically map all 16 endpoints.

Yes, that is the primary advantage of this framework. When the server returns data from `search_studies`, Pydantic AI ensures the response strictly matches your defined Python models before passing it to the LLM.

The framework throws a validation error immediately. If `search_by_location` returns malformed JSON or unexpected fields, the agent stops execution rather than hallucinating an answer.

Your agent calls the `search_recruiting` tool. It returns only the trials currently accepting patients, which you can then filter further by condition or sponsor.

The integration only handles public registry data, specifically NCT IDs, phase classifications, and eligibility criteria. Because the MCP environment operates on ephemeral infrastructure, your exact Pydantic schemas and search parameters are destroyed immediately after the response is delivered.

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