Guides

Overview

The b.well Health SDK for AI enables large language model (LLM) applications to access b.well' through the Model Context Protocol (MCP). Using MCP, your AI applications connect to a b.well-hosted MCP server and gain access to a catalog of tools (agents). These tools allow LLMs to access b.well services, capabilities, and health data within your application's AI workflows

Choose the Health SDK for AI when you:

  • Build AI-powered healthcare applications with LLMs
  • Need to integrate patient health data into conversational AI workflows
  • Want to leverage pre-built healthcare agents without custom API integration
  • Are using MCP-compatible platforms (OpenAI, Anthropic, Google)

Integration At a Glance

  • Your AI application uses an MCP client to connect an LLM to the b.well-hosted MCP server (ex. OpenAI, Anthropic, or Google)
  • The b.well MCP server exposes a catalog of b.well MCP tools the model can call as needed
  • Each request is made on behalf of an authorized end user (see End-User Authentication)
  • The MCP server executes the request and returns structured results that the model can use

MCP Client Configuration Guides

b.well's MCP server works with MCP-compatible clients. All platforms access the same b.well tools and capabilities—only the client configuration differs.

Authentication: All MCP requests operate in an end-user context. Authentication setup can vary by LLM platform and is covered in each configuration guide below.

Select your platform for setup instructions:

MCP Tools (Agents) Catalog

Once your LLM is configured, it can invoke b.well tools to access user health data and various b.well services.

See the Tools / Agents Catalog for the complete list of available tools, parameters, and response formats.


💡 Tip: To support various user journey workflows, the Health SDK for AI can be combined with other End-User Application integration patterns to supplement the available MCP tools (agents).


Updated 2 months ago