Added multi-agent content

Author: buzahid

learn-pr/paths/develop-ai-agents-on-azure/index.yml

@@ -30,5 +30,6 @@ modules:
 - learn.wwl.build-agent-with-custom-tools
 - learn.wwl.develop-ai-agent-with-semantic-kernel
 - learn.wwl.orchestrate-sk-multi-agent-solution
+- learn.wwl.develop-multi-agent-azure-ai-foundry
 trophy:
   uid: learn.wwl.develop-ai-agent-on-azure.trophy

learn-pr/wwl-data-ai/develop-multi-agent-azure-ai-foundry/1-introduction.yml

@@ -0,0 +1,13 @@
+### YamlMime:ModuleUnit
+uid: learn.wwl.develop-multi-agent-azure-ai-foundry.introduction
+title: Introduction
+metadata:
+  title: Introduction
+  description: "Get started with multi-agent systems using Azure AI Foundry Agent Service."
+  ms.date: 06/13/2025
+  author: buzahid
+  ms.author: buzahid
+  ms.topic: unit
+durationInMinutes: 2
+content: |
+  [!include[](includes/1-introduction.md)]

learn-pr/wwl-data-ai/develop-multi-agent-azure-ai-foundry/2-understand-connected-agents.yml

@@ -0,0 +1,13 @@
+### YamlMime:ModuleUnit
+uid: learn.wwl.develop-multi-agent-azure-ai-foundry.understand-connected-agents
+title: Understand connected agents
+metadata:
+  title: Understand connected agents
+  description: "Learn how connected agents work together in a cohesive multi-agent solution"
+  ms.date: 06/13/2025
+  author: buzahid
+  ms.author: buzahid
+  ms.topic: unit
+durationInMinutes: 5
+content: |
+  [!include[](includes/2-understand-connected-agents.md)]

learn-pr/wwl-data-ai/develop-multi-agent-azure-ai-foundry/3-design-multi-agent-solution.yml

@@ -0,0 +1,13 @@
+### YamlMime:ModuleUnit
+uid: learn.wwl.develop-multi-agent-azure-ai-foundry.design-multi-agent-solution
+title: Design a multi-agent solution with connected agents
+metadata:
+  title: Design a multi-agent solution with connected agents
+  description: "Learn how to create a multi-agent solution using Azure AI Foundry Agent Service."
+  ms.date: 06/13/2025
+  author: buzahid
+  ms.author: buzahid
+  ms.topic: unit
+durationInMinutes: 5
+content: |
+  [!include[](includes/3-design-multi-agent-solution.md)]

learn-pr/wwl-data-ai/develop-multi-agent-azure-ai-foundry/4-exercise.yml

@@ -0,0 +1,13 @@
+### YamlMime:ModuleUnit
+uid: learn.wwl.develop-multi-agent-azure-ai-foundry.exercise
+title: Exercise - Develop a multi-agent app with Azure AI Foundry
+metadata:
+  title: Exercise - Develop a multi-agent app with Azure AI Foundry
+  description: "Build a multi-agent solution using connected agents in Azure AI Foundry Agent Service."
+  ms.date: 06/13/2025
+  author: buzahid
+  ms.author: buzahid
+  ms.topic: unit
+durationInMinutes: 30
+content: |
+  [!include[](includes/4-exercise.md)]

learn-pr/wwl-data-ai/develop-multi-agent-azure-ai-foundry/5-knowledge-check.yml

@@ -0,0 +1,47 @@
+### YamlMime:ModuleUnit
+uid: learn.wwl.develop-multi-agent-azure-ai-foundry.knowledge-check
+title: Module assessment
+metadata:
+  title: Module assessment
+  description: "Check your learning on connected agents in Azure AI Foundry Agent Service."
+  ms.date: 06/13/2025
+  author: buzahid
+  ms.author: buzahid
+  ms.topic: unit
+durationInMinutes: 3
+content: |
+quiz:
+  questions:
+  - content: "What is the role of the main agent in a connected agent system?"
+    choices:
+    - content: "To directly perform all tasks using tools"
+      isCorrect: false
+      explanation: "Incorrect. The main agent delegates tasks rather than performing them directly."
+    - content: "To coordinate user input and route tasks to the appropriate connected agents"
+      isCorrect: true
+      explanation: "Correct. The main agent handles intent recognition and delegates work to connected agents."
+    - content: "To monitor agent performance and generate logs"
+      isCorrect: false
+      explanation: "Incorrect. Logging is not the role of the main agent."
+  - content: "How do you connect an agent to a main agent using the Azure AI Projects client library?"
+    choices:
+    - content: "Add the agent as a `ConnectedAgentTool` to the main agent's tool definition."
+      isCorrect: true
+      explanation: "Correct. You register the agent as a tool and include it when configuring the main agent."
+    - content: "Use the `link_agents()` method to bind the sub-agent to the main agent."
+      isCorrect: false
+      explanation: "Incorrect. There is no such method; connected agents are registered as tools."
+    - content: "Set the main agent's parent_id to the sub-agent's agent ID."
+      isCorrect: false
+      explanation: "Incorrect. Agent relationships are not configured through ID inheritance."
+  - content: "How does the main agent decide which connected agent to use?"
+    choices:
+    - content: "It uses prompt instructions and natural language understanding."
+      isCorrect: true
+      explanation: "Correct. The main agent uses natural language prompts to determine task routing."
+    - content: "It follows a fixed code-based decision tree."
+      isCorrect: false
+      explanation: "Incorrect. No hardcoded routing logic is needed."
+    - content: "It randomly selects from available connected agents."
+      isCorrect: false
+      explanation: "Incorrect. Task routing is based on the user's intent and the prompt design."

learn-pr/wwl-data-ai/develop-multi-agent-azure-ai-foundry/6-summary.yml

@@ -0,0 +1,13 @@
+### YamlMime:ModuleUnit
+uid: learn.wwl.develop-multi-agent-azure-ai-foundry.summary
+title: Summary
+metadata:
+  title: Summary
+  description: "Reflect on what you've learned about multi-agent systems."
+  ms.date: 06/13/2025
+  author: buzahid
+  ms.author: buzahid
+  ms.topic: unit
+durationInMinutes: 1
+content: |
+  [!include[](includes/6-summary.md)]

learn-pr/wwl-data-ai/develop-multi-agent-azure-ai-foundry/includes/1-introduction.md

@@ -0,0 +1,5 @@
+Azure AI Foundry Agent Service can help you develop sophisticated, multi-agent systems that can break down complex tasks into smaller, specialized roles. Using connected agents, you can design intelligent solutions where a primary agent delegates work to sub-agents without the need for custom orchestration logic or hardcoded routing. This modular approach boosts efficiency and maintainability across a wide range of scenarios.
+
+Imagine you're on an engineering team that receives a constant flow of support tickets including bugs, feature requests, and infrastructure issues. Manually reviewing and sorting each one takes time and slows down your team’s ability to respond quickly. With a multi-agent approach, you can build a triage assistant that assigns different tasks to specialized agents. The sub-agents might perform tasks such as classifying ticket type, setting priority, and suggesting the right team for the work. With a multi-agent approach, these specialized agents can work together to streamline the ticketing process.
+
+In this module, you'll learn how to use connected agents in Azure AI Foundry. You'll also practice building an intelligent ticket triage system using a collaborative mutli-agent solution.

learn-pr/wwl-data-ai/develop-multi-agent-azure-ai-foundry/includes/2-understand-connected-agents.md

@@ -0,0 +1,31 @@
+As AI solutions become more advanced, managing complex workflows gets harder. A single agent can handle a wide range of tasks, but this approach can become unmanageable as the scope expands. That's why Azure AI Foundry Agent Service lets you connect multiple agents, each with a focused role, to work together in a cohesive system.
+
+## What are connected agents?
+
+Connected agents are a feature in the Azure AI Foundry Agent Service that allow you to break large tasks into smaller,specialized roles without building a custom orchestrator or hardcoding routing logic. Instead of relying on one agent to do everything, you can create multiple agents with clearly defined responsibilities that collaborate to accomplish tasks.
+
+At the center of this system is a main agent that interprets user input and delegates tasks to connected sub-agents. Each sub-agent is purpose-built to perform a specific function, such as summarizing a document, validating a policy, or retrieving data from a knowledge source.
+
+This division of labor helps you:
+
+- Simplify complex workflows
+- Improve agent performance and accuracy
+- Make systems easier to maintain and extend over time
+
+## Why use connected agents?
+
+Rather than scaling a single agent to handle every user request or data interaction, using connected agents lets you:
+
+- Build modular solutions that are easier to develop and debug
+- Assign specialized capabilities to agents that can be reused across solutions
+- Scale your system in a way that aligns with real-world business logic
+
+This approach is especially useful in scenarios where agents need to perform sensitive tasks independently, such has handling private data or generating personalized content.
+
+Using connected agents to automate workflows offers many benefits, for example:
+
+- **No custom orchestration required** - The main agent uses natural language to route tasks, eliminating the need for hardcoded logic.
+- **Improved reliability and traceability** - The clear separation of responsibilities makes issues easier to debug since agents can be tested individually.
+- **Flexible and extensible** - Add or swap agents without reworking the entire system or modifying the main agent.
+
+Connected agents make it easier to build modular, collaborative systems without complex orchestration. By assigning focused roles and using natural language delegation, you can simplify workflows, improve reliability, and scale your solutions more effectively.

learn-pr/wwl-data-ai/develop-multi-agent-azure-ai-foundry/includes/3-design-multi-agent-solution.md

@@ -0,0 +1,54 @@
+In a connected agent solution, success depends on clearly defining the responsibilities of each agent. The central agent is also responsible for how the agents will collaborate. Let's explore how to design a multi-agent program using Azure AI Foundry Agent Service.
+
+**Main agent responsibilities**
+
+The main agent acts as the orchestrator. It interprets the intent behind a request and determines which connected agent is best suited to handle it. The main agent is responsible for:
+
+- Interpreting user input
+- Selecting the appropriate connected agent
+- Forwarding relevant context and instructions
+- Aggregating or summarize results
+
+**Connected agent responsibilities**
+
+Connected agents designed to focus on a single domain of responsibility. A connected agent is responsible for:
+
+- Completing a specific action based on a clear prompt
+- Using tools (if needed) to complete their task
+- Returning the results to the main agent
+
+Connected agents should be designed with a single responsibility in mind. This makes your system easier to debug, extend, and reuse.
+
+## Set up a multi-agent solution with connected agents
+
+1. **Initialize the agents client**
+
+    First, you create a client that will connect to your Azure AI Foundry project.
+
+1. **Create an agent to connect to the main agent**
+
+   Define an agent that will be connected to the main agent. You can do this using the `create_agent` method on the `AgentsClient` object. 
+   
+   For example, your connected agent might retrieve stock prices, summarize documents, or validate compliance. Give the agent clear instructions that define its purpose.
+
+1. **Initialize the connected agent tool**
+
+    Use your agent definition to create a `ConnectedAgentTool`. Assign it a name and description so the main agent knows when and how to use it.
+
+1. **Create the main agent**
+
+    Create the main agent using the `create_agent` method. Add your connected agents using the `tools` property and assign the `ConnectedAgentTool` definitions to the main agent.
+
+1. **Create a thread and send a message**
+
+    Create the agent thread that will be used to manage the conversation context. Then create a message on the thread that contains the request you want the agent to fulfill.
+
+1. **Run the agent workflow**
+
+    Once the message is added, create a run to process the request. The main agent uses its tools to delegate tasks as needed and compile a final response for the user.
+
+1. **Handle the results**
+
+    When the run completes, you can review the main agent's response. The final output may incorporate insights from one or more connected agents. Only the main agent's response is visible to the end user.
+
+Designing a connected agent system involves defining focused agents, registering them as tools, and configuring a main agent to route tasks intelligently. This modular approach gives you a flexible foundation for building collaborative AI solution that scale as your needs grow.