product management agent with demo task delegation

Author: brown9804

README.md

@@ -1,4 +1,4 @@
-# Demo: Zava AI Shopping Assistant <br/> Multi-Agent Architecture with A2A Protocol - Overview
+# Demo: Zava AI Shopping Assistant <br/> Multi-Agent Architecture with A2A Protocol - Overview 
 
 Costa Rica
 
@@ -10,7 +10,7 @@ Last updated: 2025-12-03
 ----------
 
 > [!IMPORTANT]
-> Disclaimer: This repository contains a demo of `Zava AI Shopping Assistant`, a  multi-agent system designed for e-commerce. It features a fully automated `"Zero-Touch" deployment` pipeline orchestrated by Terraform, which `provisions infrastructure, ingests data, creates real AI agents in MSFT Foundry, and deploys the application container.` Please refer [TechWorkshop L300: AI Apps and Agents](https://microsoft.github.io/TechWorkshop-L300-AI-Apps-and-agents/), and if needed contact Microsoft directly: [Microsoft Sales and Support](https://support.microsoft.com/contactus?ContactUsExperienceEntryPointAssetId=S.HP.SMC-HOME) more guindace. There are tons of free resources out there, all eager to support!
+> Disclaimer: This repository contains a demo of `Zava AI Shopping Assistant`, an enhanced multi-agent system implementing Agent-to-Agent (A2A) protocol for e-commerce. It features a fully automated `"Zero-Touch" deployment` pipeline orchestrated by Terraform, which `provisions infrastructure, ingests data, creates specialized AI agents with delegation patterns in MSFT Foundry, and deploys the complete A2A application stack.` Please refer [TechWorkshop L300: AI Apps and Agents](https://microsoft.github.io/TechWorkshop-L300-AI-Apps-and-agents/), and if needed contact Microsoft directly: [Microsoft Sales and Support](https://support.microsoft.com/contactus?ContactUsExperienceEntryPointAssetId=S.HP.SMC-HOME) for more guidance. There are tons of free resources out there, all eager to support!
 
 <img width="1905" height="1086" alt="image" src="https://github.com/user-attachments/assets/5cd2776f-4606-45c2-9482-53ff2d4df74e" />
 
@@ -23,16 +23,20 @@ Last updated: 2025-12-03
 
 ## Key Features
 
-- **A2A Protocol Implementation**: Complete Agent-to-Agent communication framework with standardized messaging, event handling, and task coordination
-- **Multi-Agent Architecture**: Specialized AI agents working through A2A protocol:
-  - **Cora (Shopper)**: Front-facing assistant for general queries
-  - **Inventory Manager**: Checks stock availability via A2A requests
-  - **Customer Loyalty**: Manages rewards and discounts through agent coordination
-  - **Cart Manager**: Handles shopping cart operations with inter-agent communication
-- **Real Azure AI Agents**: Integrates with **MSFT Foundry** to create and host persistent agents (not just local simulations)
-- **Zero-Touch Deployment**: A single [terraform apply](./terraform-infrastructure/README.md) command handles the entire lifecycle including A2A framework deployment
-- **A2A Intelligent Routing**: Enhanced Handoff Service that supports both traditional routing and A2A protocol agent discovery
-- **Data Pipeline Automation**: Automatically ingests product catalogs with A2A event notifications and coordination
+- **Enhanced A2A Protocol**: Agent-to-Agent communication with delegation patterns, specialized agent coordination, and factual data integration
+- **6-Agent Architecture**: Specialized AI agents with proper delegation through A2A protocol:
+  - **Cora (Shopper)**: Front-facing assistant for general customer queries
+  - **Interior Design Specialist**: Design expertise and style recommendations  
+  - **Inventory Manager**: Stock availability and product lookup coordination
+  - **Customer Loyalty**: Rewards management and discount optimization
+  - **Cart Manager**: Shopping cart operations and checkout coordination
+  - **Product Management Specialist**: Coordinates with Marketing Agent, Ranker Agent, and Product Information Plugin for comprehensive product services
+- **Specialized Agent Delegation**: Product Manager delegates marketing tasks to Marketing Agent and ranking tasks to Ranker Agent as appropriate
+- **Factual Data Plugin**: Product Information Plugin provides accurate product catalog data from predefined sources
+- **Real MSFT Foundry Agents**: Integrates with **MSFT Foundry** to create and host persistent agents with proper delegation patterns
+- **Zero-Touch Deployment**: A single [terraform apply](./terraform-infrastructure/README.md) command handles the entire lifecycle including enhanced A2A framework deployment
+- **A2A Task Coordination**: Advanced inter-agent task delegation with specialized expertise routing
+- **Data Pipeline Automation**: Automatically ingests product catalogs with comprehensive A2A event coordination
 
 ## About A2A Protocol
 
@@ -49,11 +53,13 @@ Last updated: 2025-12-03
 > A2A Components in This Project:
 
 - **Agent Execution Framework**: Manages multiple agent instances (`src/a2a/server/agent_execution.py`)
-- **Event Queue System**: Handles inter-agent communication (`src/a2a/server/events/`)
-- **Task Management**: Coordinates work between agents (`src/a2a/server/tasks.py`)
-- **Request Handlers**: Processes agent-to-agent requests (`src/a2a/server/request_handlers.py`)
-- **Coordinator Agent**: Orchestrates multi-agent workflows (`src/a2a/agent/coordinator.py`)
-- **API Endpoints**: RESTful and WebSocket APIs for agent communication (`src/a2a/api/`)
+- **Event System**: Handles inter-agent communication and delegation (`src/a2a/server/events/`)
+- **Task Coordination**: Advanced task delegation between specialized agents (`src/a2a/server/tasks.py`)
+- **Request Handlers**: Processes agent-to-agent requests with delegation routing (`src/a2a/server/request_handlers.py`)
+- **Coordinator Agent**: Orchestrates complex multi-agent workflows (`src/a2a/agent/coordinator.py`)
+- **Specialized Agents**: Marketing Agent, Ranker Agent with delegation patterns (`src/app/agents/`)
+- **Product Information Plugin**: Factual data source for product catalog (`src/app/agents/product_information_plugin.py`)
+- **API Endpoints**: RESTful and WebSocket APIs for enhanced agent communication (`src/a2a/api/`)
 
 > A2A vs Traditional Multi-Agent Systems:
 
@@ -79,17 +85,26 @@ graph TD
     Router -->|Inventory Events| Inventory[Inventory Agent]
     Router -->|Loyalty Tasks| Loyalty[Loyalty Agent]
     Router -->|Cart Events| Cart[Cart Agent]
+    Router -->|Product Tasks| ProductMgr[Product Manager]
+    
+    ProductMgr -->|Marketing Tasks| Marketing[Marketing Agent]
+    ProductMgr -->|Ranking Tasks| Ranker[Ranker Agent]
+    ProductMgr -->|Factual Data| Plugin[Product Info Plugin]
     
     subgraph "A2A Communication"
         EventQueue <--> Cora
         EventQueue <--> Design
         EventQueue <--> Inventory
         EventQueue <--> Loyalty
         EventQueue <--> Cart
+        EventQueue <--> ProductMgr
+        EventQueue <--> Marketing
+        EventQueue <--> Ranker
     end
     
     Inventory -->|Query| Search[Azure AI Search]
     Inventory -->|Lookup| Cosmos[Cosmos DB]
+    Plugin -->|Catalog| PredefinedData[Product Catalog Data]
 ```
 
 ## What Happens Under the Hood?
@@ -119,12 +134,16 @@ graph TD
 
         <https://github.com/user-attachments/assets/37c4a8cd-73e1-4392-8755-fb018481d8cb>
 
-4. **Agent Creation & A2A Registration**:
-   - Installs the `azure-ai-projects` SDK.
-   - Connects to MSFT Foundry.
-   - Provisions 5 real agents with A2A protocol integration and specific instructions.
-   - Registers agents with the A2A discovery service.
-   - Saves the unique Agent IDs and A2A endpoints to the `.env` file.
+4. **Enhanced Agent Creation & A2A Registration**:
+   - Installs the `azure-ai-projects` SDK and Microsoft Agent Framework.
+   - Connects to MSFT Foundry for agent hosting.
+   - Provisions 6 specialized agents with enhanced A2A protocol integration:
+     - Core shopping agents (5) plus Product Management Specialist
+     - Marketing Agent and Ranker Agent with delegation patterns
+     - Product Information Plugin with predefined catalog data
+   - Registers all agents with the enhanced A2A discovery service.
+   - Configures delegation relationships between Product Manager and specialized agents.
+   - Saves the unique Agent IDs, delegation endpoints, and A2A configuration to the `.env` file.
 
       <img width="1907" height="990" alt="image" src="https://github.com/user-attachments/assets/4234ead4-16e0-4bec-bbfd-3b8891d113b0" />
 
@@ -149,18 +168,23 @@ graph TD
    - Check A2A Server API: `https://<your-app-name>.azurewebsites.net/a2a/api/docs`
    - Verify agent discovery: `https://<your-app-name>.azurewebsites.net/a2a/server/agents`
 
-3. **Verify Agents**:
+3. **Verify Enhanced Agent Architecture**:
    - Go to the [MSFT Foundry Portal](https://ai.azure.com).
    - Navigate to your project -> **Build** -> **Agents**.
-   - You should see all 5 agents listed with A2A protocol integration.
+   - You should see all 6 agents listed with enhanced A2A protocol integration:
+     - Core agents: Cora, Interior Design, Inventory, Loyalty, Cart Manager
+     - Product Management Specialist with delegation capabilities
 
       <https://github.com/user-attachments/assets/3c562ccd-cff3-4a30-b9f8-44111fb71113>
 
-4. **Test A2A Interactions**: For example:
+4. **Test Enhanced A2A Interactions**: For example:
    - **General**: "Hi, who are you?" (Handled by Cora via A2A protocol)
    - **Inventory**: "Do you have the classic leather sofa in stock?" (Routed through A2A to Inventory Agent)
    - **Design**: "What colors of green paint do you have?" (A2A task delegation to Design Agent)
-   - **Multi-Agent**: "Find a sofa and check my loyalty points" (A2A coordination between multiple agents)
+   - **Product Recommendations**: "Recommend modern furniture for my living room" (Product Manager delegates to Marketing Agent)
+   - **Product Comparisons**: "Compare sectional sofas" (Product Manager delegates to Ranker Agent)
+   - **Product Details**: "What are the specifications of product SOFA-001?" (Product Manager uses Product Information Plugin)
+   - **Multi-Agent**: "Find a sofa, check reviews, and verify my loyalty points" (Complex A2A coordination across multiple specialized agents)
 
 <!-- START BADGE -->
 <div align="center">

src/a2a/.env_automation

@@ -4,11 +4,11 @@ A2A_PORT=8001
 A2A_LOG_LEVEL=INFO
 
 # Base application URL for monitoring
-BASE_APP_URL=https://zava-72910920-app.azurewebsites.net
+BASE_APP_URL=https://zava-51c07969-app.azurewebsites.net
 
 # Azure monitoring integration
-APPLICATION_INSIGHTS_CONNECTION_STRING=InstrumentationKey=cd157009-2ed7-472b-9bcf-9f83189fe438;IngestionEndpoint=https://westus3-1.in.applicationinsights.azure.com/;LiveEndpoint=https://westus3.livediagnostics.monitor.azure.com/;ApplicationId=43df942d-375b-4d95-b0b4-a85f1045018c
-LOG_ANALYTICS_WORKSPACE_ID=9a4604f5-a37e-4fc6-9c14-73d1d63e88d7
+APPLICATION_INSIGHTS_CONNECTION_STRING=InstrumentationKey=3ecb7f2d-bf9b-4c15-973f-39e820e44b10;IngestionEndpoint=https://westus3-1.in.applicationinsights.azure.com/;LiveEndpoint=https://westus3.livediagnostics.monitor.azure.com/;ApplicationId=c47cdc80-ebeb-4b09-a7f5-4f9b6c9502dc
+LOG_ANALYTICS_WORKSPACE_ID=d3733137-0fd4-48ea-ad35-e3d066ec4d0b
 
 # Automation features
 ENABLE_PROCESS_MANAGEMENT=true

src/a2a/agent/__init__.py

@@ -6,7 +6,8 @@
     ZavaAgentAdapter, InteriorDesignAgentAdapter, InventoryAgentAdapter,
     CustomerLoyaltyAgentAdapter, CartManagementAgentAdapter, CoraAgentAdapter
 )
-from .coordinator import A2ACoordinatorAgent, EnhancedProductManagementAgent
+from .coordinator import A2ACoordinatorAgent, EnhancedProductManagementAgent as CoordinatorEnhancedAgent
+from .product_management_agent import EnaganecedProductManagementAgent
 
 __all__ = [
     "ZavaAgentAdapter",
@@ -16,5 +17,5 @@
     "CartManagementAgentAdapter", 
     "CoraAgentAdapter",
     "A2ACoordinatorAgent",
-    "EnhancedProductManagementAgent"
+    "EnaganecedProductManagementAgent"
 ]

src/a2a/agent/coordinator.py

@@ -27,6 +27,7 @@
     CartManagementAgentAdapter, CoraAgentAdapter
 )
 
+
 # Import existing handoff service
 import sys
 import os
@@ -197,7 +198,7 @@ def _simple_classification(self, user_message: str) -> Dict[str, Any]:
             best_domain = max(scores, key=scores.get)
             confidence = min(0.8, scores[best_domain] * 0.2)  # Max 0.8 confidence
         else:
-            best_domain = "cora"  # Default fallback
+            best_domain = "product_management"  # Default to product management
             confidence = 0.3
 
         return {
@@ -215,8 +216,8 @@ async def _route_to_agent(
     ) -> None:
         """Route request to the appropriate agent"""
         if domain not in self.agents:
-            logger.warning(f"Unknown domain: {domain}, falling back to cora")
-            domain = "cora"
+            logger.warning(f"Unknown domain: {domain}, falling back to product_management")
+            domain = "product_management"
 
         target_agent = self.agents[domain]
         self.active_handoffs[task.id] = domain

src/a2a/main.py

@@ -37,7 +37,7 @@
 from a2a.api import A2AChatRouter, A2AServerRouter
 from a2a.api.enhanced_chat_router import EnhancedA2AChatRouter
 from a2a.server import A2AStarletteApplication, DefaultRequestHandler
-from a2a.agent import EnhancedProductManagementAgent
+from a2a.agent import EnaganecedProductManagementAgent
 from a2a.types import AgentCard, AgentCapabilities, AgentSkill
 
 # Import legacy components for hybrid mode

src/a2a/status_automation.ps1

@@ -11,7 +11,7 @@ if () {
 
 # Check automation endpoint
 try {
-   = Invoke-RestMethod -Uri "https://zava-72910920-app.azurewebsites.net/a2a/automation/status" -TimeoutSec 5
+   = Invoke-RestMethod -Uri "https://zava-51c07969-app.azurewebsites.net/a2a/automation/status" -TimeoutSec 5
   Write-Host "Automation Status: "
 } catch {
   Write-Host "Automation endpoint not accessible"