Merge pull request #53524 from buzahid/ai-agents

AnnaMHuff · web-flow · commit 4be5123565a6 · 2026-02-18T09:51:21.000-07:00
Update AI Agents modules
diff --git a/learn-pr/wwl-data-ai/build-agent-workflows-microsoft-foundry/1-introduction.yml b/learn-pr/wwl-data-ai/build-agent-workflows-microsoft-foundry/1-introduction.yml
@@ -3,10 +3,10 @@ uid: learn.wwl.build-agent-workflows-microsoft-foundry.introduction
 title: Introduction
 metadata:
   title: Introduction
-  description: Get started with building agent-driven workflows using Microsoft Foundry.
+  description: Get started with building multi-agent workflows using Microsoft Foundry.
   author: buzahid
   ms.author: buzahid
-  ms.date: 1/8/2026
+  ms.date: 2/18/2026
   ms.topic: unit
 durationInMinutes: 1
 content: |
diff --git a/learn-pr/wwl-data-ai/build-agent-workflows-microsoft-foundry/includes/1-introduction.md b/learn-pr/wwl-data-ai/build-agent-workflows-microsoft-foundry/includes/1-introduction.md
@@ -2,7 +2,7 @@ Modern AI solutions often rely on multiple agents working together to analyze in
 
 Foundry includes a visual workflow builder that lets you design and test these systems without writing extensive code. Using the canvas, you can define how agents are invoked, how data moves between steps, and how decisions are made based on agent outputs. You can also observe execution paths and inspect intermediate results to understand how your workflow behaves at runtime.
 
-Suppose you're a developer responsible for automating customer support workflows at a growing SaaS company. Your team receives a steady stream of support tickets ranging from billing disputes to API errors and simple how-to questions. Manually reviewing each request doesn't scale, but fully automating responses isn't always safe. Workflows allow you to combine AI agents, conditional logic, and human-in-the-loop escalation. By using agent-driven workflows, you can triage multiple support requests efficiently and at scale while maintaining reliability and control.
+Suppose you're a developer responsible for automating customer support workflows at a growing SaaS company. Your team receives a steady stream of support tickets ranging from billing disputes to API errors and simple how-to questions. Manually reviewing each request doesn't scale, but fully automating responses isn't always safe. Workflows allow you to combine multiple AI agents, conditional logic, and human-in-the-loop escalation. By using agent-driven workflows, you can triage multiple support requests efficiently and at scale while maintaining reliability and control.
 
 ## After completing this module, you'll be able to:
 
diff --git a/learn-pr/wwl-data-ai/build-agent-workflows-microsoft-foundry/index.yml b/learn-pr/wwl-data-ai/build-agent-workflows-microsoft-foundry/index.yml
@@ -5,7 +5,7 @@ metadata:
   description: Workflows enable you to orchestrate AI agents and other components to create intelligent applications. Learn how to build and manage workflows using Microsoft Foundry.
   author: buzahid
   ms.author: buzahid
-  ms.date: 1/8/2026
+  ms.date: 2/18/2026
   ms.service: azure-ai-foundry
   ms.topic: module-standard-task-based
   ai-usage: ai-assisted
diff --git a/learn-pr/wwl-data-ai/connect-agent-to-mcp-tools/1-introduction.yml b/learn-pr/wwl-data-ai/connect-agent-to-mcp-tools/1-introduction.yml
@@ -4,7 +4,7 @@ title: Introduction
 metadata:
   title: Introduction
   description: "Learn how to integrate Model Context Protocol (MCP) tools with Azure AI Agents."
-  ms.date: 06/13/2025
+  ms.date: 2/18/2026
   author: buzahid
   ms.author: buzahid
   ms.topic: unit
diff --git a/learn-pr/wwl-data-ai/connect-agent-to-mcp-tools/2-understand-mcp-tool-discovery.yml b/learn-pr/wwl-data-ai/connect-agent-to-mcp-tools/2-understand-mcp-tool-discovery.yml
@@ -4,7 +4,7 @@ title: Understand MCP tool discovery
 metadata:
   title: Understand MCP Tool Discovery
   description: "Learn how Model Context Protocol (MCP) servers host functions that can be discovered by LLMs."
-  ms.date: 06/13/2025
+  ms.date: 2/18/2026
   author: buzahid
   ms.author: buzahid
   ms.topic: unit
diff --git a/learn-pr/wwl-data-ai/connect-agent-to-mcp-tools/3-mcp-client-server-setup.yml b/learn-pr/wwl-data-ai/connect-agent-to-mcp-tools/3-mcp-client-server-setup.yml
@@ -4,7 +4,7 @@ title: Integrate agent tools using an MCP server and client
 metadata:
   title: Integrate Agent Tools Using an MCP Server and Client
   description: "Learn how to expose Model Context Protocol (MCP) tools through a server and register them to an Azure AI Agent."
-  ms.date: 06/13/2025
+  ms.date: 2/18/2026
   author: buzahid
   ms.author: buzahid
   ms.topic: unit
diff --git a/learn-pr/wwl-data-ai/connect-agent-to-mcp-tools/4-use-azure-ai-agents-with-mcp.yml b/learn-pr/wwl-data-ai/connect-agent-to-mcp-tools/4-use-azure-ai-agents-with-mcp.yml
@@ -4,7 +4,7 @@ title: Use Azure AI agents with MCP servers
 metadata:
   title: Use Azure AI agents with MCP servers
   description: "Learn how to connect your Azure AI Agent to remote Model Context Protocol (MCP) servers"
-  ms.date: 07/21/2025
+  ms.date: 2/18/2026
   author: buzahid
   ms.author: buzahid
   ms.topic: unit
diff --git a/learn-pr/wwl-data-ai/connect-agent-to-mcp-tools/5-exercise.yml b/learn-pr/wwl-data-ai/connect-agent-to-mcp-tools/5-exercise.yml
@@ -4,7 +4,7 @@ title: Exercise - Connect MCP tools to Azure AI Agents
 metadata:
   title: Exercise - Connect MCP Tools to Azure AI Agents
   description: "Build a Model Context Protocol (MCP) client-server application that dynamically registers tools to an Azure AI Agent."
-  ms.date: 07/21/2025
+  ms.date: 2/18/2026
   author: buzahid
   ms.author: buzahid
   ms.topic: unit
diff --git a/learn-pr/wwl-data-ai/connect-agent-to-mcp-tools/6-knowledge-check.yml b/learn-pr/wwl-data-ai/connect-agent-to-mcp-tools/6-knowledge-check.yml
@@ -4,7 +4,7 @@ title: Module assessment
 metadata:
   title: Module Assessment
   description: "Check your learning on connecting Model Context Protocol (MCP) tools to Azure AI Agents."
-  ms.date: 06/13/2025
+  ms.date: 2/18/2026
   author: buzahid
   ms.author: buzahid
   ms.topic: unit
diff --git a/learn-pr/wwl-data-ai/connect-agent-to-mcp-tools/7-summary.yml b/learn-pr/wwl-data-ai/connect-agent-to-mcp-tools/7-summary.yml
@@ -4,7 +4,7 @@ title: Summary
 metadata:
   title: Summary
   description: "Reflect on what you learned about Model Context Protocol (MCP) agent tools."
-  ms.date: 06/13/2025
+  ms.date: 2/18/2026
   author: buzahid
   ms.author: buzahid
   ms.topic: unit
diff --git a/learn-pr/wwl-data-ai/connect-agent-to-mcp-tools/includes/4-use-azure-ai-agents-with-mcp.md b/learn-pr/wwl-data-ai/connect-agent-to-mcp-tools/includes/4-use-azure-ai-agents-with-mcp.md
@@ -8,34 +8,29 @@ To connect to an MCP server, you need:
 - A remote MCP server endpoint (for example, https://api.githubcopilot.com/mcp/).
 - A Microsoft Foundry agent configured to use the MCP tool.
 
-You can connect to multiple MCP servers by adding them as separate tools, each with:
+You can connect to multiple MCP servers by adding them to your agent as separate tools. Each `MCPTool` can include the following parameters:
 - `server_label`: A unique identifier for the MCP server (e.g., GitHub).
 - `server_url`: The MCP server’s URL.
 - `allowed_tools` (optional): A list of specific tools the agent is allowed to access.
+- `require_approval` (optional): A boolean that determines whether tool invocations require human approval. If set to true, the agent will pause and wait for approval before invoking any tools on the MCP server.
 
 The MCP tool also supports custom headers, which let you pass:
 - Authentication keys (API keys, OAuth tokens).
 - Other required headers for the MCP server.
 
-    - These headers are included in `tool_resources` during each run and are not stored between runs.
-
 ## Invoking tools
 
 When using the Azure MCP Tool object, you don't need to wrap function tools or invoke `session.call_tool`. Instead, the tools are automatically invoked when necessary during an agent run. To automatically invoke MCP tools:
 
-- Create the `McpTool` object with the server label and url.
+- Create the `MCPTool` object with the server label and url.
 - Use `update_headers` to apply any headers required by the server.
-- Use the `set_approval_mode` to  determine whether approval is required. Supported values are:
+- Use the `require_approval` parameter to determine whether approval is required. Supported values are:
     - `always`: A developer needs to provide approval for every call. If you don't provide a value, this one is the default.
     - `never`: No approval is required.
-- Create a `ToolSet` object and add the `McpTool` object
-- Create an agent run and specify the `toolset` property
-- When the run completes, you should see the results of any invoked tools in the response.
-
-If the model tries to invoke a tool in your MCP server with approval required, you get a run status of requires_action. 
-    - In the `requires_action` field, you can get more details on which tool in the MCP server is called and any arguments to be passed.
-    - Review the tool and arguments so that you can make an informed decision for approval.
-    - Submit your approval to the agent with call_id by setting approve to true.
+- Create an agent and add the `MCPTool` object to its tools list
+- Invoke a prompt on the agent, you should see the results of any invoked tools in the response.
+
+If the model tries to invoke a tool in your MCP server with approval required, you get an `mcp_approval_request` in the agent response. This includes information about which tool is being invoked, and you can use this information to decide whether to approve the request. To approve, you send a follow-up message with the `mcp_approval_response` object, which includes an `approval_request_id` value and an `approve` boolean.
 
 MCP integration is a key step toward creating richer, more context-aware AI agents. As the MCP ecosystem grows, you’ll have even more opportunities to bring specialized tools into your workflows and deliver smarter, more dynamic solutions.
 
diff --git a/learn-pr/wwl-data-ai/connect-agent-to-mcp-tools/includes/5-exercise.md b/learn-pr/wwl-data-ai/connect-agent-to-mcp-tools/includes/5-exercise.md
@@ -5,4 +5,4 @@ If you have an Azure subscription, you can complete this exercise to develop a M
 
 Launch the exercise and follow the instructions.
 
-[![Button to launch exercise.](../media/launch-exercise.png)](https://go.microsoft.com/fwlink/?linkid=2325472&azure-portal=true)
+[![Button to launch exercise.](../media/launch-exercise.png)](https://go.microsoft.com/fwlink/?linkid=2353400&azure-portal=true)
diff --git a/learn-pr/wwl-data-ai/connect-agent-to-mcp-tools/includes/7-summary.md b/learn-pr/wwl-data-ai/connect-agent-to-mcp-tools/includes/7-summary.md
@@ -3,5 +3,6 @@ In this module, you learned how to integrate external tools with Microsoft Found
 By connecting your agent to an MCP server, you can dynamically discover and register tools at runtime without hardcoding APIs or redeploying your agent. Using an MCP client, you generated function wrappers from discovered tools and connected them directly to your agent. This integration allows your agent to adapt to evolving toolsets, and create more flexible AI solutions that can grow alongside your applications.
 
 > [!TIP]
-> To learn more about MCP, see the **[Model Context Protocol User Guide](https://modelcontextprotocol.io/introduction)** and **[AI Agents MCP Integration](https://github.com/microsoft/ai-agents-for-beginners/tree/main/11-agentic-protocols#readme)**.
-> To learn more about using Microsoft Foundry Agent Service with MCP, visit **[Connect to Model Context Protocol servers](/azure/ai-foundry/agents/how-to/tools/model-context-protocol)**.
+> To learn more about MCP, see the **[Model Context Protocol User Guide](https://modelcontextprotocol.io/introduction)**.
+>
+> To learn more about using Microsoft Foundry Agent Service with MCP, visit **[Connect to Model Context Protocol servers](/azure/ai-foundry/agents/how-to/tools/model-context-protocol?view=foundry&preserve-view=true&pivots=python)**.
diff --git a/learn-pr/wwl-data-ai/connect-agent-to-mcp-tools/index.yml b/learn-pr/wwl-data-ai/connect-agent-to-mcp-tools/index.yml
@@ -3,7 +3,7 @@ uid: learn.wwl.connect-agent-to-mcp-tools
 metadata:
   title: Integrate MCP Tools with Azure AI Agents
   description: Enable dynamic tool access for your Azure AI agents. Learn how to connect MCP-hosted tools and integrate them seamlessly into agent workflows.
-  ms.date: 08/28/2025
+  ms.date: 2/18/2026
   author: buzahid
   ms.author: buzahid
   ms.topic: module-standard-task-based # Please don't edit, used for our analytics
