Merge pull request #304917 from cephalin/ai-python

add stack tabs

Author: v-ccolin

articles/app-service/overview-ai-integration.md

@@ -5,16 +5,17 @@ author: cephalin
 ms.author: cephalin
 ms.service: azure-app-service
 ms.topic: overview
-ms.date: 06/12/2025
+ms.date: 09/01/2025
 ms.custom:
   - build-2025
 ---
 
 # Integrate AI into your Azure App Service applications
 
+
 This article guides you to language-specific tutorials and resources to help you build intelligent applications with App Service.
 
-Azure App Service makes it easy to integrate AI capabilities into your web applications across multiple programming languages and frameworks. Whether you want to use powerful Azure OpenAI models, deploy local small language models (SLMs) directly with your apps, host Model Context Protocol (MCP) servers or implement advanced patterns like retrieval augmented generation (RAG), App Service provides the flexible, secure platform you need for AI-powered applications.
+Azure App Service makes it easy to integrate AI capabilities into your web applications across multiple programming languages and frameworks. Whether you want to use powerful Azure OpenAI models, deploy local small language models (SLMs) directly with your apps, build agentic web applications, expose your app as a tool for AI agents using OpenAPI, host Model Context Protocol (MCP) servers, or implement advanced patterns like retrieval augmented generation (RAG), App Service provides the flexible, secure platform you need for AI-powered applications.
 
 App Service offers several advantages for developing and deploying AI-powered applications:
 
@@ -23,67 +24,103 @@ App Service offers several advantages for developing and deploying AI-powered ap
 - **Enterprise-grade security** - Implement network isolation, end-to-end encryption, and role-based access control
 - **Simplified DevOps with GitHub integration** - Streamline CI/CD pipelines using GitHub Actions, leverage GitHub Codespaces with integrated GitHub Copilot for AI-assisted development, and create end-to-end workflows from development to production deployment
 
-## .NET applications
+## Build chatbots and RAG applications in App Service
 
-Build AI-powered .NET applications with these tutorials:
+Build intelligent web apps that use Azure OpenAI for chat or retrieval augmented generation (RAG). These tutorials show you how to integrate Azure OpenAI and (optionally) Azure AI Search to create chatbots and RAG solutions in your preferred language, using managed identities for secure authentication.
 
-- [Build a chatbot with Azure OpenAI (Blazor)](tutorial-ai-openai-chatbot-dotnet.md) - Create a Blazor web app that connects to Azure OpenAI to generate TLDR summaries using Semantic Kernel.
-- [Build a RAG application with Azure OpenAI and Azure AI Search (.NET)](tutorial-ai-openai-search-dotnet.md) - Implement RAG to enable your AI models to access and use your organization's data.
-- [Integrate an App Service app as an MCP Server for GitHub Copilot Chat (.NET)](tutorial-ai-model-context-protocol-server-dotnet.md)
-- [Add an App Service app as a tool in Azure AI Foundry Agent Service (.NET)](tutorial-ai-integrate-azure-ai-agent-dotnet.md)
+### [.NET](#tab/dotnet)
+- [Build a chatbot with Azure OpenAI (Blazor)](tutorial-ai-openai-chatbot-dotnet.md)
+- [Build a RAG application with Azure OpenAI and Azure AI Search (.NET)](tutorial-ai-openai-search-dotnet.md)
+- [Build a RAG application with Azure OpenAI and Azure SQL](deploy-intelligent-apps-dotnet-to-azure-sql.md)
+
+### [Java](#tab/java)
+- [Build a chatbot with Azure OpenAI (Spring Boot)](tutorial-ai-openai-chatbot-java.md)
+- [Build a RAG application with Azure OpenAI and Azure AI Search (Java)](tutorial-ai-openai-search-java.md)
+- Sample: [SpringBoot-Petclinic-AI-Chat-on-App-Service](https://github.com/Azure-Samples/SpringBoot-Petclinic-AI-Chat-on-App-Service)
+
+### [Node.js](#tab/nodejs)
+- [Build a chatbot with Azure OpenAI (Express.js)](tutorial-ai-openai-chatbot-node.md)
+- [Build a RAG application with Azure OpenAI and Azure AI Search (Node.js)](tutorial-ai-openai-search-nodejs.md)
+
+### [Python](#tab/python)
+- [Build a chatbot with Azure OpenAI (Flask)](tutorial-ai-openai-chatbot-python.md)
+- [Build a RAG application with Azure OpenAI and Azure AI Search (Python)](tutorial-ai-openai-search-python.md)
+- [Azure AI Foundry tutorial: Deploy an enterprise chat web app](/azure/ai-foundry/tutorials/deploy-chat-web-app?toc=/azure/app-service/toc.json&bc=/azure/bread/toc.json)
+-----
+
+## Build agentic web applications
+
+Transform your traditional CRUD web applications for the AI era by adding agentic capabilities with frameworks like Microsoft Semantic Kernel, LangGraph, or Azure AI Foundry Agent Service. Instead of users navigating forms, textboxes, and dropdowns, you can offer a conversational interface that lets users "talk to an agent" that intelligently performs the same operations your app provides. This approach enables your web app to reason, plan, and take actions on behalf of users.
+
+### [.NET](#tab/dotnet)
 - [Tutorial: Build an agentic web app in Azure App Service with Microsoft Semantic Kernel or Azure AI Foundry Agent Service (.NET)](tutorial-ai-agent-web-app-semantic-kernel-foundry-dotnet.md)
-- [Build a RAG application with Azure OpenAI and Azure SQL](deploy-intelligent-apps-dotnet-to-azure-sql.md) - Use Azure SQL as a vector database for RAG applications.
-- [Run a chatbot with a local SLM sidecar extension](tutorial-ai-slm-dotnet.md) - Deploy a chatbot that uses a local SLM without requiring an external AI service.
-- [Invoke a web app from Azure AI Foundry Agent](invoke-openapi-web-app-from-azure-ai-agent-service.md) - Make your web API available to AI agents.
-- [Build an agentic web app with Semantic Kernel Agent2Agent (A2A) Protocol integration](https://techcommunity.microsoft.com/blog/appsonazureblog/building-agent-to-agent-a2a-applications-on-azure-app-service/4433114) - Deploy a multi-agent system where a main agent coordinates with specialized agents using [A2A](https://a2aproject.github.io/A2A/latest/).
 
-## Java applications
+### [Java](#tab/java)
+- [Tutorial: Build an agentic web app in Azure App Service with Microsoft Semantic Kernel (Spring Boot)](tutorial-ai-agent-web-app-semantic-kernel-java.md)
 
-Integrate AI capabilities into your Java applications:
+### [Node.js](#tab/nodejs)
+- [Tutorial: Build an agentic web app in Azure App Service with LangGraph or Azure AI Foundry Agent Service (Node.js)](tutorial-ai-agent-web-app-langgraph-foundry-node.md)
 
-- [Build a chatbot with Azure OpenAI (Spring Boot)](tutorial-ai-openai-chatbot-java.md) - Create a Spring Boot application that connects to Azure OpenAI using managed identity.
-- [Build a RAG application with Azure OpenAI and Azure AI Search (Java)](tutorial-ai-openai-search-java.md) - Implement RAG to search through your own documents with Java.
-- [Run a chatbot with a local SLM (Spring Boot)](tutorial-ai-slm-spring-boot.md) - Deploy a Spring Boot application with a local SLM sidecar.
-- [Integrate an App Service app as an MCP Server for GitHub Copilot Chat (Java)](tutorial-ai-model-context-protocol-server-java.md) - Host an MCP server with Java to extend GitHub Copilot Chat capabilities.
-- [Add an App Service app as a tool in Azure AI Foundry Agent Service (Java)](tutorial-ai-integrate-azure-ai-agent-java.md) - Make your Java application available as a tool for AI agents.
-- [Build an agentic web app with Semantic Kernel (Java)](tutorial-ai-agent-web-app-semantic-kernel-java.md) - Create an intelligent agent-based application using Java and Semantic Kernel.
+### [Python](#tab/python)
+- [Tutorial: Build an agentic web app in Azure App Service with LangGraph or Azure AI Foundry Agent Service (Python)](tutorial-ai-agent-web-app-langgraph-foundry-python.md)
+-----
 
-Samples:
+## App Service as OpenAPI tool in Azure AI Foundry Agent
 
-- [SpringBoot-Petclinic-AI-Chat-on-App-Service](https://github.com/Azure-Samples/SpringBoot-Petclinic-AI-Chat-on-App-Service)
+Empower your existing web apps by exposing their capabilities to Azure AI Foundry Agent Service using OpenAPI. Many web apps already provide REST APIs, making them ideal candidates for integration into agents that can call REST APIs as tools. By connecting Azure AI Foundry Agent Service to these APIs, you can rapidly create powerful, feature-rich agents with little code.
+
+### [.NET](#tab/dotnet)
+- [Add an App Service app as a tool in Azure AI Foundry Agent Service (.NET)](tutorial-ai-integrate-azure-ai-agent-dotnet.md)
+- [Invoke a web app from Azure AI Foundry Agent](invoke-openapi-web-app-from-azure-ai-agent-service.md)
+
+### [Java](#tab/java)
+- [Add an App Service app as a tool in Azure AI Foundry Agent Service (Java)](tutorial-ai-integrate-azure-ai-agent-java.md)
+
+### [Node.js](#tab/nodejs)
+- [Add an App Service app as a tool in Azure AI Foundry Agent Service (Node.js)](tutorial-ai-integrate-azure-ai-agent-node.md)
+
+### [Python](#tab/python)
+- [Add an App Service app as a tool in Azure AI Foundry Agent Service (Python)](tutorial-ai-integrate-azure-ai-agent-python.md)
+-----
+
+## App Service as Model Context Protocol (MCP) servers
+
+Integrate your web app as a [Model Context Protocol (MCP)](https://modelcontextprotocol.io/introduction) server to extend the capabilities of leading personal AI agents such as GitHub Copilot Chat, Cursor, and Winsurf. By exposing your app's APIs through MCP, you can supercharge these agents with the unique features and business logic your web app already provides, without major development effort or rearchitecture.
+
+### [.NET](#tab/dotnet)
+- [Integrate an App Service app as an MCP Server for GitHub Copilot Chat (.NET)](tutorial-ai-model-context-protocol-server-dotnet.md)
+- Sample: [Host a .NET MCP server on Azure App Service](https://github.com/Azure-Samples/remote-mcp-webapp-dotnet)
 
-## Node.js applications
+### [Java](#tab/java)
+- [Integrate an App Service app as an MCP Server for GitHub Copilot Chat (Java)](tutorial-ai-model-context-protocol-server-java.md)
 
-Add AI features to your Node.js web applications:
+### [Node.js](#tab/nodejs)
+- [Integrate an App Service app as an MCP Server for GitHub Copilot Chat (Node.js)](tutorial-ai-model-context-protocol-server-node.md)
+- Sample: [Host a Node.js MCP server on Azure App Service](https://github.com/Azure-Samples/remote-mcp-webapp-node)
 
-- [Build a chatbot with Azure OpenAI (Express.js)](tutorial-ai-openai-chatbot-node.md) - Create an Express.js application that connects to Azure OpenAI using managed identity.
-- [Build a RAG application with Azure OpenAI and Azure AI Search (Node.js)](tutorial-ai-openai-search-nodejs.md) - Build a RAG application with Node.js.
-- [Run a chatbot with a local SLM (Express.js)](tutorial-ai-slm-expressjs.md) - Deploy an Express.js application with a local SLM sidecar.
-- [Integrate an App Service app as an MCP Server for GitHub Copilot Chat (Node.js)](tutorial-ai-model-context-protocol-server-node.md) - Host an MCP server with Node.js to extend GitHub Copilot Chat capabilities.
-- [Add an App Service app as a tool in Azure AI Foundry Agent Service (Node.js)](tutorial-ai-integrate-azure-ai-agent-node.md) - Make your Node.js application available as a tool for AI agents.
-- [Build an agentic web app with LangGraph and Azure AI Foundry (Node.js)](tutorial-ai-agent-web-app-langgraph-foundry-node.md) - Create an intelligent agent-based application using Node.js and a choicer of LangGraph or Azure AI Foundry Agent Service.
+### [Python](#tab/python)
+- [Integrate an App Service app as an MCP Server for GitHub Copilot Chat (Python)](tutorial-ai-model-context-protocol-server-python.md)
+- Sample: [Host a Python MCP server on Azure App Service](https://github.com/Azure-Samples/remote-mcp-webapp-python)
+- Sample: [Host a Python MCP server with key-based authorization on Azure App Service](https://github.com/Azure-Samples/remote-mcp-webapp-python-auth)
+- Sample: [Host a Python MCP server with OAuth 2.0 authorization on Azure App Service](https://github.com/Azure-Samples/remote-mcp-webapp-python-auth-oauth) - Deploy an MCP server with Python and [Open Authorization (OAuth) 2.0 authorization with Microsoft Entra ID](/entra/architecture/auth-oauth2).
+-----
 
-## Python applications
+## Use a local SLM (sidecar container)
 
-Implement AI capabilities in your Python web applications:
+Deploy a web app with a local small language model (SLM) as a sidecar container to run AI models entirely within your App Service environment. No outbound calls or external AI service dependencies required. This approach is ideal if you have strict data privacy or compliance requirements, as all AI processing and data remain local to your app. App Service offers high-performance, memory-optimized pricing tiers needed for running SLMs in sidecars.
 
-- [Build a chatbot with Azure OpenAI (Flask)](tutorial-ai-openai-chatbot-python.md) - Create a Flask application that connects to Azure OpenAI using managed identity.
-- [Build a RAG application with Azure OpenAI and Azure AI Search (Python)](tutorial-ai-openai-search-python.md) - Implement RAG with Python.
-- [Run a chatbot with a local SLM (FastAPI)](tutorial-ai-slm-fastapi.md) - Deploy a FastAPI application with a local SLM sidecar.
-- [Azure AI Foundry tutorial: Deploy an enterprise chat web app](/azure/ai-foundry/tutorials/deploy-chat-web-app?toc=/azure/app-service/toc.json&bc=/azure/bread/toc.json) - Deploy fully integrated AI web app straight from your deployment in Azure AI Foundry.
-- [Integrate an App Service app as an MCP Server for GitHub Copilot Chat (Python)](tutorial-ai-model-context-protocol-server-python.md) - Host an MCP server with Python to extend GitHub Copilot Chat capabilities.
-- [Add an App Service app as a tool in Azure AI Foundry Agent Service (Python)](tutorial-ai-integrate-azure-ai-agent-python.md) - Expose your Python web app’s capabilities to Azure AI Foundry Agent Service using OpenAPI, enabling agents to perform real-world tasks.
-- [Build an agentic web app with LangGraph or Azure AI Foundry Agent Service (Python)](tutorial-ai-agent-web-app-langgraph-foundry-python.md) - Create an intelligent agent-based application using Python, LangGraph, or Azure AI Foundry Agent Service.
+### [.NET](#tab/dotnet)
+- [Run a chatbot with a local SLM sidecar extension](tutorial-ai-slm-dotnet.md)
 
-## Model Context Protocol (MCP) servers
+### [Java](#tab/java)
+- [Run a chatbot with a local SLM (Spring Boot)](tutorial-ai-slm-spring-boot.md)
 
-Host [Model Context Protocol (MCP)](https://modelcontextprotocol.io/introduction) in your web applications:
+### [Node.js](#tab/nodejs)
+- [Run a chatbot with a local SLM (Express.js)](tutorial-ai-slm-expressjs.md)
 
-- [Host a .NET MCP server on Azure App Service](https://github.com/Azure-Samples/remote-mcp-webapp-dotnet) - Deploy an MCP server with .NET.
-- [Host a Node.js MCP server on Azure App Service](https://github.com/Azure-Samples/remote-mcp-webapp-node) - Deploy an MCP server with Node.js.
-- [Host a Python MCP server on Azure App Service](https://github.com/Azure-Samples/remote-mcp-webapp-python) - Deploy an MCP server with Python.
-- [Host a Python MCP server with key-based authorization on Azure App Service](https://github.com/Azure-Samples/remote-mcp-webapp-python-auth) - Deploy an MCP server with Python and key-based authorization.
-- [Host a Python MCP server with OAuth 2.0 authorization on Azure App Service](https://github.com/Azure-Samples/remote-mcp-webapp-python-auth-oauth) - Deploy an MCP server with Python and [Open Authorization (OAuth) 2.0 authorization with Microsoft Entra ID](/entra/architecture/auth-oauth2).
+### [Python](#tab/python)
+- [Run a chatbot with a local SLM (FastAPI)](tutorial-ai-slm-fastapi.md)
+-----
 
 ## More resources