How to Use the Epic Fhir MCP in LlamaIndex

Q: How do I load Epic Fhir resources into a LlamaIndex vector store?

Connect using the MCP client and wrap it in McpToolSpec. Call totoollistasync() to fetch tools like listdiagnosticreports, then index the resulting JSON strings directly into your document store.

Q: Can I limit which Epic Fhir tools my LlamaIndex agent can run?

Yes, use the allowedtools filter when defining your MCP tool specification. This lets you restrict the agent to non-sensitive endpoints like listappointments while blocking access to clinical records.

Q: How does LlamaIndex handle raw JSON payloads from Epic Fhir?

The framework ingests outputs from tools like listobservations as node documents. It parses the values, units, and reference ranges to construct a structured index for semantic queries.

Q: Does this integration support asynchronous tool execution?

Yes, you can use totoollistasync() to load and execute tools concurrently. This is critical when fetching large datasets like listimmunizations and listencounters at the same time.

Q: How does this server secure patient demographics and clinical observations?

Data fetched via searchpatients and listobservations travels through an isolated V8 sandbox that decrypts payloads in transit and destroys them immediately after. Your credentials and patient records never touch persistent disk storage.

Index live Epic Fhir data directly into LlamaIndex vector stores for hallucination-free medical RAG.

See Vinkius in Action

Works with every AI agent you already use

…and any MCP-compatible client

MCP Servers - Free for Subscribers

Connect Epic Fhir MCP to LlamaIndex

Create your Vinkius account to connect Epic Fhir to LlamaIndex 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 Epic Fhir with LlamaIndex

Ask AI about this MCP

ChatGPT

Claude

Perplexity

Build patient-grounded RAG pipelines

This MCP Server exposes tools like `list_diagnostic_reports` and `list_observations` to feed your LlamaIndex knowledge bases. Instead of guessing clinical values, your agent runs queries against live EHR data to ground its responses in verified clinical facts. You index the returned JSON payloads directly into your local vector store. This mechanism guarantees that any synthesis of a patient's history is backed by real-time lab results and vitals.

Index Epic Fhir tool outputs into LlamaIndex

Running `list_conditions` and `list_medications` allows your agent to pull structured patient histories. LlamaIndex takes this raw tool output and indexes it, turning discrete FHIR resources into searchable, semantic documents. You can search past queries and clinical summaries using standard semantic search. This approach bridges the gap between structured API endpoints and unstructured clinical notes.

Control tool access with strict filters

Using `list_appointments` and `list_encounters` with `McpToolSpec` lets you restrict which endpoints your agent can touch during a session. If a workflow only requires reading schedules, you use the `allowed_tools` filter to only expose these calendar tools. You pass these filtered tools to your `FunctionAgent` to limit the agent's action space. This safety measure ensures the model never attempts unauthorized queries on broader patient datasets.

Setup guide

Set up Epic Fhir MCP in LlamaIndex

Prerequisites

Python 3.10+ installed
llama-index-tools-mcp package
Active Vinkius subscription with a valid endpoint token

1

Install dependencies
Run pip install llama-index-tools-mcp llama-index-llms-openai. The MCP tools package provides BasicMCPClient and McpToolSpec.
2

Connect with BasicMCPClient
Point BasicMCPClient to your Vinkius endpoint URL. Replace [YOUR_TOKEN_HERE] with your token from cloud.vinkius.com. Supports SSE and Streamable HTTP transports.
3

Convert to LlamaIndex tools
Call mcp_tool_spec.to_tool_list_async() to convert all Epic Fhir MCP tools into native FunctionTool objects that any LlamaIndex agent can use.
4

Run with any LLM
Create a FunctionAgent with the tools and your preferred LLM. Swap OpenAI for Anthropic, Gemini, or any LlamaIndex-supported provider.

agent.py

from llama_index.tools.mcp import BasicMCPClient, McpToolSpec
from llama_index.core.agent.workflow import FunctionAgent
from llama_index.llms.openai import OpenAI

# Connect to the MCP
mcp_client = BasicMCPClient(
    "https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp"
)
mcp_tool_spec = McpToolSpec(client=mcp_client)

# Convert MCP tools to LlamaIndex tools
tools = await mcp_tool_spec.to_tool_list_async()

# Create and run the agent
agent = FunctionAgent(
    tools=tools,
    llm=OpenAI(model="gpt-4o"),
    system_prompt="You have access to Epic Fhir tools.",
)
response = await agent.run("List recent Epic Fhir data")

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 Epic Fhir. 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 Epic Fhir 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 Epic Fhir MCP in LlamaIndex

Connect using the MCP client and wrap it in `McpToolSpec`. Call `to_tool_list_async()` to fetch tools like `list_diagnostic_reports`, then index the resulting JSON strings directly into your document store.

Yes, use the `allowed_tools` filter when defining your MCP tool specification. This lets you restrict the agent to non-sensitive endpoints like `list_appointments` while blocking access to clinical records.

The framework ingests outputs from tools like `list_observations` as node documents. It parses the values, units, and reference ranges to construct a structured index for semantic queries.

Yes, you can use `to_tool_list_async()` to load and execute tools concurrently. This is critical when fetching large datasets like `list_immunizations` and `list_encounters` at the same time.

Data fetched via `search_patients` and `list_observations` travels through an isolated V8 sandbox that decrypts payloads in transit and destroys them immediately after. Your credentials and patient records never touch persistent disk storage.

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