Configure an MCP client, give it access to a patient via the four VFS tools, and walk through a real clinical question end-to-end.

Connect an AI Agent

In this tutorial you will connect a language model to Patient Memory using the Model Context Protocol (MCP). By the end, the agent will be able to browse a patient's conditions, read a condition story, and answer a clinical question that requires traversing relationships across the graph.

Time: ~20 minutes.

Prerequisites

A patient already ingested. Complete Ingest and Query a Patient first.
Node.js 18+.
An Anthropic API key and a Patient Memory access token set in your environment:

export ANTHROPIC_API_KEY="your-anthropic-key"
export CLINIA_ACCESS_TOKEN="your-access-token"   # obtained via OAuth client credentials

See Manage Credentials to obtain a Patient Memory access token.

Replace <workspace-id> with your Patient Memory workspace ID in the code below.

Steps

Install dependencies

npm install @modelcontextprotocol/sdk @anthropic-ai/sdk

Connect the MCP client

Patient Memory exposes an MCP endpoint at /mcp over HTTP with SSE streaming. Connect to it and confirm the four tools are available:

import { Client } from "@modelcontextprotocol/sdk/client/index.js";
import { SSEClientTransport } from "@modelcontextprotocol/sdk/client/sse.js";

const transport = new SSEClientTransport(new URL("https://<workspace-id>.w.clinia.cloud/mcp"), {
  requestInit: {
    headers: { Authorization: `Bearer ${process.env.CLINIA_ACCESS_TOKEN}` },
  },
});

const mcp = new Client({ name: "clinical-agent", version: "1.0.0" });
await mcp.connect(transport);

const { tools } = await mcp.listTools();
console.log(tools.map((t) => t.name));
// → ["browse_patient", "read_patient", "search_patient", "get_patient_info"]

The {id} segment in VFS paths is the VFS patient ID — the patient.id extracted from the ingested record. It may differ from the registry key used at ingest time. Use get_patient_info to look it up if needed.

Wire the tools to Claude

Pass the MCP tools directly to the Anthropic SDK and let Claude decide when to call them:

import Anthropic from "@anthropic-ai/sdk";

const anthropic = new Anthropic();

const mcpTools = tools.map((t) => ({
  name: t.name,
  description: t.description ?? "",
  input_schema: t.inputSchema,
}));

async function runAgent(question: string): Promise<string> {
  const messages: Anthropic.MessageParam[] = [{ role: "user", content: question }];

  while (true) {
    const response = await anthropic.messages.create({
      model: "claude-sonnet-4-6",
      max_tokens: 4096,
      tools: mcpTools,
      messages,
    });

    if (response.stop_reason === "end_turn") {
      return response.content
        .filter((b) => b.type === "text")
        .map((b) => b.text)
        .join("");
    }

    const toolUses = response.content.filter((b) => b.type === "tool_use");
    if (toolUses.length === 0) break;

    const toolResults = await Promise.all(
      toolUses.map(async (use) => {
        if (use.type !== "tool_use") return null;
        const result = await mcp.callTool({
          name: use.name,
          arguments: use.input as Record<string, unknown>,
        });
        return {
          type: "tool_result" as const,
          tool_use_id: use.id,
          content: result.content
            .filter((c) => c.type === "text")
            .map((c) => c.text)
            .join("\n"),
        };
      }),
    );

    messages.push({ role: "assistant", content: response.content });
    messages.push({
      role: "user",
      content: toolResults.filter(Boolean) as Anthropic.ToolResultBlockParam[],
    });
  }

  return "";
}

Run a clinical question

const answer = await runAgent(
  "For patient jeanne-tremblay: given the recent acute COPD exacerbation, " +
    "is the current inhaler regimen appropriate and is bone health being monitored?",
);

console.log(answer);
await mcp.close();

The agent typically calls tools in this sequence:

browse_patient on /patient/jeanne-72f-copd/conditions/active to discover available conditions
read_patient on chronic_obstructive_lung_disease/_story.md — the COPD story includes the acute exacerbation, current inhaler regimen, and spirometry trends
Optionally read_patient on osteopenia/_story.md — surfaces the ICS exposure as a contributing factor and the pending DEXA follow-up
Synthesises the answer from the graph data

The agent reaches this conclusion without any application code that knows about ICS-related bone loss. It discovers the relationship from the condition stories, which were assembled by the pipeline from the clinical knowledge base.

Next steps

MCP Tools reference for full parameter documentation
Virtual File System to understand the path schema your agent is navigating
Read a condition story for a focused how-to on format and token budget choices

Connect an AI Agent

On this page