Skip to content

README.md

Latest commit

 

History

History
114 lines (88 loc) · 6.93 KB

README.md

File metadata and controls

114 lines (88 loc) · 6.93 KB

Zava Insurance — Declarative Agent with MCP Server & Rich UI

A Microsoft 365 Copilot Declarative Agent that connects to the Zava Insurance MCP Server, enabling insurance claims management through natural language. The MCP server uses the OpenAI App SDK to render rich, interactive widgets directly inside the Copilot chat — including claims dashboards, claim detail views, and contractor lists.

Watch the demo

Watch the demo on YouTube | Demo video file

Built with the Agents Toolkit (ATK) in VS Code. Instead of hand-authoring an OpenAPI spec, ATK points at the MCP discovery URL and generates all manifests, wiring in tools and function definitions automatically.

What This Agent Can Do

Rich UI Tools (render interactive widgets in chat)

Tool Description
show-claims-dashboard Grid view of all claims with status filters, metrics, and click-to-detail
show-claim-detail Detailed view of a single claim with inspections, purchase orders, and a map
show-contractors Filterable list of contractors with ratings and specialties

Data Tools

Tool Description
update-claim-status Update a claim's status and add notes
update-inspection Update inspection status, findings, and recommended actions
update-purchase-order Update a purchase order's status
get-claim-summary Text summary of a specific claim
list-inspectors List all inspectors with specializations

Sample Prompts

Prompt What it does
Show the claims dashboard Opens the claims dashboard widget with all claims, status metrics, and click-to-detail
Show me all open claims sorted by estimated loss from highest to lowest Opens the dashboard filtered to open claims and sorted by estimated loss descending — quickly surfaces the highest-value open claims
Show me Kimberly King's claim details, and tell me what inspections are pending Fetches the claim detail widget for the specific policy holder and summarizes pending inspection status — saves toggling between screens
Show me the preferred roofing contractors Opens the contractors list filtered to preferred roofing specialists — useful when assigning repair work on storm or roof damage claims
Approve claim 2 with a note that all documentation has been verified, then show me the updated dashboard Updates the claim status to Approved, adds a note, and re-opens the dashboard so you can confirm the change — a multi-step workflow in one prompt
Create a high-priority initial inspection for claim CN202504990 scheduled for next Monday, and assign it to an inspector who specializes in fire damage Lists inspectors, picks one with fire damage specialization, and creates the inspection — chains three tools automatically
Which claims have the highest estimated losses? Show me the top ones and compare their damage types Opens the dashboard sorted by estimated loss descending, then the AI analyzes damage types across high-value claims to surface patterns
Show the claim detail for claim 1. Then approve the pending purchase order and mark the inspection as completed with findings noting that all repairs are satisfactory Chains claim detail view, purchase order approval, and inspection update in one conversation — replaces multiple manual steps

Prerequisites

Getting Started

  1. Create a file .env.dev file (use the sample .env.dev.sample) inside the env folder in the root of the project.

  2. Run the setup commands:

    Run all scripts from src/mcpserver/

    1. Install dependencies — run npm run install:all
    2. Start Azurite (local storage emulator) — npm run start:azurite in a separate terminal
    3. Seed the database — npm run seed
    4. Build widgets — npm run build:widgets
    5. Start the MCP server — npm run dev:server (runs on http://localhost:3001/mcp)

  3. Create a dev tunnel — Use Dev Tunnels to expose your local MCP server publicly:

    devtunnel host -p 3001 --allow-anonymous

    Copy the forwarded URL (e.g. https://<tunnel-id>.devtunnels.ms) and update the url field under the RemoteMCPServer runtime in appPackage/ai-plugin.json:

     "runtimes": [
	 {
		 "type": "RemoteMCPServer",
		 "spec": {
			 "url": "<your-mcp-server-url>",
			 "mcp_tool_description": {
				 "...": "..."
			 }
		 }
	 }
 ]

  4. Inside src/mcpserver folder create your .env file — copy .env.sample:

    cp .env.sample .env

    This file contains the Azure Table Storage connection string and server port used by the MCP server. The defaults in .env.sample are pre-configured for local Azurite development.

  5. Provision — Use the Provision button from Agents Toolkit's LifeCycle panel.

Test the agent

  1. Open your browser and go to https://m365.cloud.microsoft/chat.
  2. Select your agent in the left-hand sidebar. If you don't see your agent, select All agents.
  3. Ask the agent to do something that invokes your MCP server, use above table for reference to sample prompts.
  4. Allow the agent to connect to the MCP server when prompted.
  5. The agent renders the UI widget.

Project Structure

Folder Description
appPackage/ Agent manifests — ai-plugin.json (tool definitions & auth), declarativeAgent.json (agent config), manifest.json (Teams/Outlook integration)
src/mcpserver/server/ MCP server — Express + StreamableHTTP transport, Azure Table Storage data layer
src/mcpserver/widgets/ React 18 + Fluent UI v9 widgets built as single-file HTML via OpenAI App SDK (claims dashboard, claim detail, contractors list)
src/mcpserver/db/ Seed data (JSON)
env/ Local environment files
m365agents.yml ATK lifecycle configuration

Learn More