[{"_1":2,"_424":-5,"_425":-5},"loaderData",{"_3":4,"_22":23,"_293":294},"root",{"_5":6,"_7":8},"toasts",[],"visitor",{"_9":10,"_11":12},"viewportWidth",1280,"visitorSession",{"_13":14,"_16":17,"_18":-5,"_19":20,"_21":-5},"attributes",[15],"bot","id",17495170,"membership","referenceId","0199ab24-15cd-719b-b50b-df06c33c002b","userAccount","routes/_public/mcp/servers/~namespace/~slug/_pages/_layout",{"_24":25,"_80":-5,"_81":38,"_27":82,"_210":211,"_292":191},"availableLocales",[26,38,45,52,59,66,73],{"_27":28,"_31":35,"_36":37},"mcpServer",{"_29":30,"_31":32,"_33":34},"description","Ein cloudbasierter benutzerdefinierter MCP-Server mit Azure Functions, der das Speichern und Abrufen von Codeausschnitten mit sicherer Kommunikation über Schlüssel, HTTPS, OAuth und Netzwerkisolationsoptionen ermöglicht.","name","Remote-MCP-Server","readme","# Erste Schritte mit Remote-MCP-Servern mithilfe von Azure Functions (Node.js/TypeScript)\n\nDies ist eine Schnellstartvorlage zum einfachen Erstellen und Bereitstellen eines benutzerdefinierten Remote-MCP-Servers in der Cloud mithilfe von Azure-Funktionen. Sie können den Server auf Ihrem lokalen Rechner klonen, wiederherstellen und mit Debugging ausführen. `azd up` können Sie ihn in wenigen Minuten in der Cloud installieren. Der MCP-Server ist standardmäßig durch Schlüssel und HTTPs geschützt und bietet zusätzliche Optionen für OAuth mit EasyAuth und/oder API Management sowie Netzwerkisolation mit VNET.\n\n**Sehen Sie sich die Videoübersicht an**\n\nWenn Sie dieses Beispiel in weiteren Sprachen suchen, sehen Sie sich die Versionen [.NET/C#](https://github.com/Azure-Samples/remote-mcp-functions-dotnet) und [Python](https://github.com/Azure-Samples/remote-mcp-functions-python) an.\n\n[![In GitHub Codespaces öffnen](https://github.com/codespaces/badge.svg)](https://codespaces.new/Azure-Samples/remote-mcp-functions-typescript)\n\nUnten sehen Sie das Architekturdiagramm für den Remote-MCP-Server mit Azure-Funktionen:\n\n![Architekturdiagramm](architecture-diagram.png)\n\n## Voraussetzungen\n\n* [Node.js](https://nodejs.org/en/download/) Version 18 oder höher\n* [Azure Functions Core Tools](https://learn.microsoft.com/azure/azure-functions/functions-run-local?pivots=programming-language-javascript#install-the-azure-functions-core-tools) >= `4.0.7030`\n* [Azure-Entwickler-CLI](https://aka.ms/azd)\n* So verwenden Sie Visual Studio Code zum lokalen Ausführen und Debuggen:\n * [Visual Studio Code](https://code.visualstudio.com/)\n * [Azure Functions-Erweiterung](https://marketplace.visualstudio.com/items?itemName=ms-azuretools.vscode-azurefunctions)\n* [Docker](https://www.docker.com/) zum Ausführen von Azurite, dem Azure Storage Emulator (optional)\n\n## Bereiten Sie Ihre lokale Umgebung vor\n\nFür dieses spezielle Beispiel wird ein Azure Storage Emulator benötigt, da wir Ausschnitte aus dem Blob-Speicher speichern und abrufen werden.\n\n1. Starten Sie Azurit\n\n ```shell\n docker run -p 10000:10000 -p 10001:10001 -p 10002:10002 \\\n mcr.microsoft.com/azure-storage/azurite\n ```\n\n> **Beachten Sie:** Wenn Sie Azurite aus der VS Code-Erweiterung verwenden, müssen Sie `Azurite: Start` , sonst werden Fehler angezeigt.\n\n## Führen Sie Ihren MCP-Server lokal vom Terminal aus aus\n\n1. Abhängigkeiten installieren\n\n ```shell\n npm install\n ```\n\n2. Erstellen des Projekts\n\n ```shell\n npm run build\n ```\n\n3. Starten Sie den Functions-Host lokal:\n\n ```shell\n func start\n ```\n\n> **Beachten Sie** , dass standardmäßig die Webhook-Route verwendet wird: `/runtime/webhooks/mcp/sse` . Später verwenden wir diese in Azure, um den Schlüssel für Client-/Host-Aufrufe festzulegen: `/runtime/webhooks/mcp/sse?code=`\n\n## Verwenden Sie den *lokalen* MCP-Server innerhalb eines Clients/Hosts\n\n### VS Code – Copilot-Bearbeitungen\n\n1. **Fügen Sie den MCP-Server** aus der Befehlspalette hinzu und fügen Sie die URL zum SSE-Endpunkt Ihrer laufenden Funktions-App hinzu:\n\n ```shell\n http://0.0.0.0:7071/runtime/webhooks/mcp/sse\n ```\n\n2. Wählen Sie **HTTP (Server-Sent-Events)** als Typ des hinzuzufügenden MCP-Servers aus.\n\n3. Geben Sie die URL zum SSE-Endpunkt Ihrer laufenden Funktions-App ein.\n\n4. Geben Sie die Server-ID ein. (Dies kann ein beliebiger Name sein.)\n\n5. Wählen Sie, ob Sie dies in Ihren Benutzereinstellungen (für alle Apps für Sie verfügbar) oder in Ihren Arbeitsbereichseinstellungen (nur für diese App verfügbar) ausführen möchten.\n\n6. **Listen Sie MCP-Server** aus der Befehlspalette auf und starten Sie den Server. Im vorherigen Schritt wurde Ihr lokaler Server möglicherweise bereits gestartet. In diesem Fall können Sie diesen Schritt überspringen.\n\n7. Geben Sie im Copilot-Chat-Agentenmodus eine Eingabeaufforderung ein, um das Tool auszulösen, z. B. wählen Sie einen Code aus und geben Sie diese Eingabeaufforderung ein\n\n ```plaintext\n Say Hello\n ```\n\n ```plaintext\n Save this snippet as snippet1 \n ```\n\n ```plaintext\n Retrieve snippet1 and apply to newFile.ts\n ```\n\n8. Wenn Sie aufgefordert werden, das Tool auszuführen, stimmen Sie zu, indem Sie auf **„Weiter“** klicken.\n\n9. Wenn Sie fertig sind, drücken Sie Strg+C im Terminalfenster, um den Hostprozess `func.exe` zu stoppen, und **listen Sie MCP-Server** aus der Befehlspalette auf und stoppen Sie den lokalen Server.\n\n### MCP-Inspektor\n\n1. Installieren und starten Sie MCP Inspector in einem **neuen Terminalfenster** .\n\n ```shell\n npx @modelcontextprotocol/inspector node build/index.js\n ```\n\n2. Wenn Sie Ihre Funktions-App zuvor gestoppt haben, starten Sie den Functions-Host lokal:\n\n ```shell\n func start\n ```\n\n3. Klicken Sie mit gedrückter STRG-Taste, um die MCP Inspector-Web-App von der von der App angezeigten URL zu laden (z. B. ).\n\n4. Stellen Sie den Transporttyp auf `SSE` ein\n\n5. Legen Sie die URL zum SSE-Endpunkt Ihrer laufenden Funktions-App fest und **stellen Sie eine Verbindung her** :\n\n ```shell\n http://0.0.0.0:7071/runtime/webhooks/mcp/sse\n ```\n\n6. **Tools auflisten** . Klicken Sie auf ein Tool und dann **auf Tool ausführen** .\n\n7. Wenn Sie fertig sind, drücken Sie Strg+C im Terminalfenster, um den Hostprozess `func.exe` zu stoppen, und drücken Sie Strg+C im Terminalfenster, um den Hostprozess `@modelcontextprotocol/inspector` zu stoppen.\n\n## Bereitstellen in Azure für Remote-MCP\n\nOptional können Sie sich für ein im Beispiel verwendetes VNet entscheiden. (Wenn Sie diese Option wählen, tun Sie dies vor `azd up` .)\n\n```bash\nazd env set VNET_ENABLED true\n```\n\nFühren Sie diesen [Azd](https://aka.ms/azd) -Befehl aus, um die Funktions-App mit allen erforderlichen Azure-Ressourcen zu versehen und Ihren Code bereitzustellen:\n\n```shell\nazd up\n```\n\n> **Beachten Sie,** dass [API Management]() zur Verbesserung der Sicherheit und Richtlinien für Ihren MCP-Server verwendet werden kann und[dass die integrierte Authentifizierung des App-Dienstes](https://learn.microsoft.com/en-us/azure/app-service/overview-authentication-authorization) zum Einrichten Ihres bevorzugten OAuth-Anbieters, einschließlich Entra, verwendet werden kann.\n\n## Stellen Sie von einem Client aus eine Verbindung mit Ihrer *Remote* -MCP-Server-Funktions-App her\n\nIhr Client benötigt einen Schlüssel, um den neuen gehosteten SSE-Endpunkt aufzurufen. Dieser hat das Format `https://.azurewebsites.net/runtime/webhooks/mcp/sse` . Die gehostete Funktion erfordert standardmäßig einen Systemschlüssel, der über das [Portal](https://learn.microsoft.com/en-us/azure/azure-functions/function-keys-how-to?tabs=azure-portal) oder die CLI abgerufen werden kann ( `az functionapp keys list --resource-group --name ` ). Rufen Sie den Systemschlüssel mit dem Namen `mcp_extension` ab.\n\n### Stellen Sie im MCP Inspector eine Verbindung zum Remote-MCP-Server her\n\nFür MCP Inspector können Sie den Schlüssel in die URL aufnehmen:\n\n```plaintext\nhttps://.azurewebsites.net/runtime/webhooks/mcp/sse?code=\n```\n\n### Verbindung zum Remote-MCP-Server in VS Code herstellen – GitHub Copilot\n\nFür GitHub Copilot in VS Code sollten Sie den Schlüssel als Header `x-functions-key` in `mcp.json` festlegen und `https://.azurewebsites.net/runtime/webhooks/mcp/sse` als URL verwenden. Das folgende Beispiel stammt aus der Datei `mcp.json` in diesem Repository und fordert Sie beim Starten des Servers in VS Code zur Eingabe des Schlüssels auf. Ihre `mcp.json` Datei sieht folgendermaßen aus:\n\n```json\n{\n \"inputs\": [\n {\n \"type\": \"promptString\",\n \"id\": \"functions-mcp-extension-system-key\",\n \"description\": \"Azure Functions MCP Extension System Key\",\n \"password\": true\n },\n {\n \"type\": \"promptString\",\n \"id\": \"functionapp-name\",\n \"description\": \"Azure Functions App Name\"\n }\n ],\n \"servers\": {\n \"remote-mcp-function\": {\n \"type\": \"sse\",\n \"url\": \"https://${input:functionapp-name}.azurewebsites.net/runtime/webhooks/mcp/sse\",\n \"headers\": {\n \"x-functions-key\": \"${input:functions-mcp-extension-system-key}\"\n }\n },\n \"local-mcp-function\": {\n \"type\": \"sse\",\n \"url\": \"http://0.0.0.0:7071/runtime/webhooks/mcp/sse\"\n }\n }\n}\n```\n\n1. Klicken Sie auf dem Server `remote-mcp-function` in der Datei `mcp.json` auf „Start“:\n\n2. Geben Sie den Namen der Funktions-App ein, die Sie im Azure-Portal erstellt haben, wenn Sie von VS Code dazu aufgefordert werden.\n\n3. Geben Sie den `Azure Functions MCP Extension System Key` in die Eingabeaufforderung ein. Sie können diesen aus dem Azure-Portal für Ihre Funktions-App kopieren, indem Sie zum Menüpunkt „Funktionen“ und dann zu „App-Schlüssel“ navigieren und den Schlüssel „ `mcp_extension` “ aus den Systemschlüsseln kopieren.\n\n4. Geben Sie im Copilot-Chat-Agentenmodus eine Eingabeaufforderung ein, um das Tool auszulösen, z. B. wählen Sie einen Code aus und geben Sie diese Eingabeaufforderung ein\n\n ```plaintext\n Say Hello\n ```\n\n ```plaintext\n Save this snippet as snippet1 \n ```\n\n ```plaintext\n Retrieve snippet1 and apply to newFile.ts\n ```\n\n## Stellen Sie Ihren Code erneut bereit\n\nSie können den Befehl `azd up` so oft ausführen, wie Sie möchten, um sowohl Ihre Azure-Ressourcen bereitzustellen als auch Codeaktualisierungen für Ihre Funktions-App bereitzustellen.\n\n> \\[!NOTE] Bereitgestellte Codedateien werden immer durch das neueste Bereitstellungspaket überschrieben.\n\n## Bereinigen von Ressourcen\n\nWenn Sie mit der Arbeit an Ihrer Funktions-App und den zugehörigen Ressourcen fertig sind, können Sie diesen Befehl verwenden, um die Funktions-App und die zugehörigen Ressourcen aus Azure zu löschen und so weitere Kosten zu vermeiden:\n\n```shell\nazd down\n```\n\n## Quellcode\n\nDer Funktionscode für die Endpunkte `getSnippet` und `saveSnippet` ist in den TypeScript-Dateien im Verzeichnis `src` definiert. Die MCP-Funktionsannotationen stellen diese Funktionen als MCP-Servertools bereit.\n\nDies zeigt den Code für einige MCP-Serverbeispiele (Zeichenfolge abrufen, Objekt abrufen, Objekt speichern):\n\n```typescript\n// Hello function - responds with hello message\nexport async function mcpToolHello(context: InvocationContext): Promise {\n return \"Hello I am MCP Tool!\";\n}\n\n// Register the hello tool\napp.mcpTool('hello', {\n toolName: 'hello',\n description: 'Simple hello world MCP Tool that responses with a hello message.',\n handler: mcpToolHello\n});\n\n// GetSnippet function - retrieves a snippet by name\nexport async function getSnippet(_message: unknown, context: InvocationContext): Promise {\n console.info('Getting snippet');\n \n // Get snippet name from the tool arguments\n const mcptoolargs = context.triggerMetadata.mcptoolargs as { snippetname?: string };\n const snippetName = mcptoolargs?.snippetname;\n\n console.info(`Snippet name: ${snippetName}`);\n \n if (!snippetName) {\n return \"No snippet name provided\";\n }\n \n // Get the content from blob binding - properly retrieving from extraInputs\n const snippetContent = context.extraInputs.get(blobInputBinding);\n \n if (!snippetContent) {\n return `Snippet '${snippetName}' not found`;\n }\n \n console.info(`Retrieved snippet: ${snippetName}`);\n return snippetContent as string;\n}\n\n\n// Register the GetSnippet tool\napp.mcpTool('getsnippet', {\n toolName: GET_SNIPPET_TOOL_NAME,\n description: GET_SNIPPET_TOOL_DESCRIPTION,\n toolProperties: [\n {\n propertyName: SNIPPET_NAME_PROPERTY_NAME,\n propertyValue: PROPERTY_TYPE,\n description: SNIPPET_NAME_PROPERTY_DESCRIPTION,\n }\n ],\n extraInputs: [blobInputBinding],\n handler: getSnippet\n});\n\n// SaveSnippet function - saves a snippet with a name\nexport async function saveSnippet(_message: unknown, context: InvocationContext): Promise {\n console.info('Saving snippet');\n \n // Get snippet name and content from the tool arguments\n const mcptoolargs = context.triggerMetadata.mcptoolargs as { \n snippetname?: string;\n snippet?: string;\n };\n \n const snippetName = mcptoolargs?.snippetname;\n const snippet = mcptoolargs?.snippet;\n \n if (!snippetName) {\n return \"No snippet name provided\";\n }\n \n if (!snippet) {\n return \"No snippet content provided\";\n }\n \n // Save the snippet to blob storage using the output binding\n context.extraOutputs.set(blobOutputBinding, snippet);\n \n console.info(`Saved snippet: ${snippetName}`);\n return snippet;\n}\n\n// Register the SaveSnippet tool\napp.mcpTool('savesnippet', {\n toolName: SAVE_SNIPPET_TOOL_NAME,\n description: SAVE_SNIPPET_TOOL_DESCRIPTION,\n toolProperties: [\n {\n propertyName: SNIPPET_NAME_PROPERTY_NAME,\n propertyValue: PROPERTY_TYPE,\n description: SNIPPET_NAME_PROPERTY_DESCRIPTION,\n },\n {\n propertyName: SNIPPET_PROPERTY_NAME,\n propertyValue: PROPERTY_TYPE,\n description: SNIPPET_PROPERTY_DESCRIPTION,\n }\n ],\n extraOutputs: [blobOutputBinding],\n handler: saveSnippet\n});\n```\n\nBeachten Sie, dass die Datei `host.json` auch einen Verweis auf das experimentelle Paket enthält, das für Apps erforderlich ist, die diese Funktion verwenden:\n\n```json\n\"extensionBundle\": {\n \"id\": \"Microsoft.Azure.Functions.ExtensionBundle.Experimental\",\n \"version\": \"[4.*, 5.0.0)\"\n}\n```\n\n## Nächste Schritte\n\n* Fügen Sie Ihrem MCP-Server [API Management](https://aka.ms/mcp-remote-apim-auth) hinzu (Authentifizierung, Gateway, Richtlinien, mehr!)\n* Fügen Sie Ihrem MCP-Server[eine integrierte Authentifizierung](https://learn.microsoft.com/en-us/azure/app-service/overview-authentication-authorization) hinzu\n* Aktivieren Sie VNET mit dem Flag VNET\\_ENABLED=true\n* Erfahren Sie mehr über [verwandte MCP-Bemühungen von Microsoft](https://github.com/microsoft/mcp/tree/main/Resources)","German","tag","de-DE",{"_27":39,"_31":43,"_36":44},{"_29":40,"_31":41,"_33":42},"A cloud-based custom MCP server using Azure Functions that enables saving and retrieving code snippets with secure communication through keys, HTTPS, OAuth, and network isolation options.","Remote MCP Server","\n# Getting Started with Remote MCP Servers using Azure Functions (Node.js/TypeScript)\n\nThis is a quickstart template to easily build and deploy a custom remote MCP server to the cloud using Azure functions. You can clone/restore/run on your local machine with debugging, and `azd up` to have it in the cloud in a couple minutes. The MCP server is secured by design using keys and HTTPs, and allows more options for OAuth using EasyAuth and/or API Management as well as network isolation using VNET. \n\n**Watch the video overview**\n\n\n $\"Watch$ \n\n\nIf you're looking for this sample in more languages check out the [.NET/C#](https://github.com/Azure-Samples/remote-mcp-functions-dotnet) and [Python](https://github.com/Azure-Samples/remote-mcp-functions-python) versions.\n\n[![Open in GitHub Codespaces](https://github.com/codespaces/badge.svg)](https://codespaces.new/Azure-Samples/remote-mcp-functions-typescript)\n\nBelow is the architecture diagram for the Remote MCP Server using Azure Functions:\n\n![Architecture Diagram](architecture-diagram.png)\n\n## Prerequisites\n\n+ [Node.js](https://nodejs.org/en/download/) version 18 or higher\n+ [Azure Functions Core Tools](https://learn.microsoft.com/azure/azure-functions/functions-run-local?pivots=programming-language-javascript#install-the-azure-functions-core-tools) >= `4.0.7030`\n+ [Azure Developer CLI](https://aka.ms/azd)\n+ To use Visual Studio Code to run and debug locally:\n + [Visual Studio Code](https://code.visualstudio.com/)\n + [Azure Functions extension](https://marketplace.visualstudio.com/items?itemName=ms-azuretools.vscode-azurefunctions)\n+ [Docker](https://www.docker.com/) to run Azurite, the Azure Storage Emulator (optional)\n\n## Prepare your local environment\n\nAn Azure Storage Emulator is needed for this particular sample because we will save and get snippets from blob storage. \n\n1. Start Azurite\n\n ```shell\n docker run -p 10000:10000 -p 10001:10001 -p 10002:10002 \\\n mcr.microsoft.com/azure-storage/azurite\n ```\n\n>**Note** if you use Azurite coming from VS Code extension you need to run `Azurite: Start` now or you will see errors.\n\n## Run your MCP Server locally from the terminal\n\n1. Install dependencies\n ```shell\n npm install\n ```\n\n1. Build the project\n ```shell\n npm run build\n ```\n\n1. Start the Functions host locally:\n ```shell\n func start\n ```\n\n> **Note** by default this will use the webhooks route: `/runtime/webhooks/mcp/sse`. Later we will use this in Azure to set the key on client/host calls: `/runtime/webhooks/mcp/sse?code=`\n\n## Use the *local* MCP server from within a client/host\n\n### VS Code - Copilot Edits\n\n1. **Add MCP Server** from command palette and add URL to your running Function app's SSE endpoint:\n ```shell\n http://0.0.0.0:7071/runtime/webhooks/mcp/sse\n ```\n\n1. Select **HTTP (Server-Sent-Events)** for the type of MCP server to add.\n1. Enter the URL to your running function app's SSE endpoint\n1. Enter the server ID. (This can be any name you want)\n1. Choose if you want to run this in your User settings (available to all apps for you) or to your Workspace settings (available to this app, only)\n1. **List MCP Servers** from command palette and start the server. The previous step may have already started your local server. If so, you can skip this step.\n1. In Copilot chat agent mode enter a prompt to trigger the tool, e.g., select some code and enter this prompt\n\n ```plaintext\n Say Hello\n ```\n\n ```plaintext\n Save this snippet as snippet1 \n ```\n\n ```plaintext\n Retrieve snippet1 and apply to newFile.ts\n ```\n1. When prompted to run the tool, consent by clicking **Continue**\n\n1. When you're done, press Ctrl+C in the terminal window to stop the `func.exe` host process, and **List MCP Servers** from command palette and stop the local server.\n\n### MCP Inspector\n\n1. In a **new terminal window**, install and run MCP Inspector\n\n ```shell\n npx @modelcontextprotocol/inspector node build/index.js\n ```\n \n1. If you stopped your function app previously, start the Functions host locally:\n\n ```shell\n func start\n ```\n\n1. CTRL click to load the MCP Inspector web app from the URL displayed by the app (e.g. http://0.0.0.0:5173/#resources)\n1. Set the transport type to `SSE` \n1. Set the URL to your running Function app's SSE endpoint and **Connect**:\n ```shell\n http://0.0.0.0:7071/runtime/webhooks/mcp/sse\n ```\n1. **List Tools**. Click on a tool and **Run Tool**. \n\n1. When you're done, press Ctrl+C in the terminal window to stop the `func.exe` host process, and press Ctrl+C in the terminal window to stop the `@modelcontextprotocol/inspector` host process.\n\n## Deploy to Azure for Remote MCP\n\nOptionally, you can opt-in to a VNet being used in the sample. (If you choose this, do this before `azd up`)\n\n```bash\nazd env set VNET_ENABLED true\n```\n\nRun this [azd](https://aka.ms/azd) command to provision the function app, with any required Azure resources, and deploy your code:\n\n```shell\nazd up\n```\n\n> **Note** [API Management]() can be used for improved security and policies over your MCP Server, and [App Service built-in authentication](https://learn.microsoft.com/en-us/azure/app-service/overview-authentication-authorization) can be used to set up your favorite OAuth provider including Entra. \n\n## Connect to your *remote* MCP server function app from a client\n\nYour client will need a key in order to invoke the new hosted SSE endpoint, which will be of the form `https://.azurewebsites.net/runtime/webhooks/mcp/sse`. The hosted function requires a system key by default which can be obtained from the [portal](https://learn.microsoft.com/en-us/azure/azure-functions/function-keys-how-to?tabs=azure-portal) or the CLI (`az functionapp keys list --resource-group --name `). Obtain the system key named `mcp_extension`.\n\n### Connect to remote MCP server in MCP Inspector\nFor MCP Inspector, you can include the key in the URL: \n```plaintext\nhttps://.azurewebsites.net/runtime/webhooks/mcp/sse?code=\n```\n\n### Connect to remote MCP server in VS Code - GitHub Copilot\nFor GitHub Copilot within VS Code, you should set the key as the `x-functions-key` header in `mcp.json`, and you would use `https://.azurewebsites.net/runtime/webhooks/mcp/sse` for the URL. The following example is from the `mcp.json` file included in this repository and uses an input to prompt you to provide the key when you start the server from VS Code. Your `mcp.json` file looks like this:\n\n```json\n{\n \"inputs\": [\n {\n \"type\": \"promptString\",\n \"id\": \"functions-mcp-extension-system-key\",\n \"description\": \"Azure Functions MCP Extension System Key\",\n \"password\": true\n },\n {\n \"type\": \"promptString\",\n \"id\": \"functionapp-name\",\n \"description\": \"Azure Functions App Name\"\n }\n ],\n \"servers\": {\n \"remote-mcp-function\": {\n \"type\": \"sse\",\n \"url\": \"https://${input:functionapp-name}.azurewebsites.net/runtime/webhooks/mcp/sse\",\n \"headers\": {\n \"x-functions-key\": \"${input:functions-mcp-extension-system-key}\"\n }\n },\n \"local-mcp-function\": {\n \"type\": \"sse\",\n \"url\": \"http://0.0.0.0:7071/runtime/webhooks/mcp/sse\"\n }\n }\n}\n```\n\n1. Click Start on the server `remote-mcp-function`, inside the `mcp.json` file:\n\n1. Enter the name of the function app that you created in the Azure Portal, when prompted by VS Code.\n\n1. Enter the `Azure Functions MCP Extension System Key` into the prompt. You can copy this from the Azure portal for your function app by going to the Functions menu item, then App Keys, and copying the `mcp_extension` key from the System Keys.\n\n1. In Copilot chat agent mode enter a prompt to trigger the tool, e.g., select some code and enter this prompt\n\n ```plaintext\n Say Hello\n ```\n\n ```plaintext\n Save this snippet as snippet1 \n ```\n\n ```plaintext\n Retrieve snippet1 and apply to newFile.ts\n ```\n\n## Redeploy your code\n\nYou can run the `azd up` command as many times as you need to both provision your Azure resources and deploy code updates to your function app.\n\n>[!NOTE]\n>Deployed code files are always overwritten by the latest deployment package.\n\n## Clean up resources\n\nWhen you're done working with your function app and related resources, you can use this command to delete the function app and its related resources from Azure and avoid incurring any further costs:\n\n```shell\nazd down\n```\n\n## Source Code\n\nThe function code for the `getSnippet` and `saveSnippet` endpoints are defined in the TypeScript files in the `src` directory. The MCP function annotations expose these functions as MCP Server tools.\n\nThis shows the code for a few MCP server examples (get string, get object, save object):\n\n```typescript\n// Hello function - responds with hello message\nexport async function mcpToolHello(context: InvocationContext): Promise {\n return \"Hello I am MCP Tool!\";\n}\n\n// Register the hello tool\napp.mcpTool('hello', {\n toolName: 'hello',\n description: 'Simple hello world MCP Tool that responses with a hello message.',\n handler: mcpToolHello\n});\n\n// GetSnippet function - retrieves a snippet by name\nexport async function getSnippet(_message: unknown, context: InvocationContext): Promise {\n console.info('Getting snippet');\n \n // Get snippet name from the tool arguments\n const mcptoolargs = context.triggerMetadata.mcptoolargs as { snippetname?: string };\n const snippetName = mcptoolargs?.snippetname;\n\n console.info(`Snippet name: ${snippetName}`);\n \n if (!snippetName) {\n return \"No snippet name provided\";\n }\n \n // Get the content from blob binding - properly retrieving from extraInputs\n const snippetContent = context.extraInputs.get(blobInputBinding);\n \n if (!snippetContent) {\n return `Snippet '${snippetName}' not found`;\n }\n \n console.info(`Retrieved snippet: ${snippetName}`);\n return snippetContent as string;\n}\n\n\n// Register the GetSnippet tool\napp.mcpTool('getsnippet', {\n toolName: GET_SNIPPET_TOOL_NAME,\n description: GET_SNIPPET_TOOL_DESCRIPTION,\n toolProperties: [\n {\n propertyName: SNIPPET_NAME_PROPERTY_NAME,\n propertyValue: PROPERTY_TYPE,\n description: SNIPPET_NAME_PROPERTY_DESCRIPTION,\n }\n ],\n extraInputs: [blobInputBinding],\n handler: getSnippet\n});\n\n// SaveSnippet function - saves a snippet with a name\nexport async function saveSnippet(_message: unknown, context: InvocationContext): Promise {\n console.info('Saving snippet');\n \n // Get snippet name and content from the tool arguments\n const mcptoolargs = context.triggerMetadata.mcptoolargs as { \n snippetname?: string;\n snippet?: string;\n };\n \n const snippetName = mcptoolargs?.snippetname;\n const snippet = mcptoolargs?.snippet;\n \n if (!snippetName) {\n return \"No snippet name provided\";\n }\n \n if (!snippet) {\n return \"No snippet content provided\";\n }\n \n // Save the snippet to blob storage using the output binding\n context.extraOutputs.set(blobOutputBinding, snippet);\n \n console.info(`Saved snippet: ${snippetName}`);\n return snippet;\n}\n\n// Register the SaveSnippet tool\napp.mcpTool('savesnippet', {\n toolName: SAVE_SNIPPET_TOOL_NAME,\n description: SAVE_SNIPPET_TOOL_DESCRIPTION,\n toolProperties: [\n {\n propertyName: SNIPPET_NAME_PROPERTY_NAME,\n propertyValue: PROPERTY_TYPE,\n description: SNIPPET_NAME_PROPERTY_DESCRIPTION,\n },\n {\n propertyName: SNIPPET_PROPERTY_NAME,\n propertyValue: PROPERTY_TYPE,\n description: SNIPPET_PROPERTY_DESCRIPTION,\n }\n ],\n extraOutputs: [blobOutputBinding],\n handler: saveSnippet\n});\n```\n\nNote that the `host.json` file also includes a reference to the experimental bundle, which is required for apps using this feature:\n\n```json\n\"extensionBundle\": {\n \"id\": \"Microsoft.Azure.Functions.ExtensionBundle.Experimental\",\n \"version\": \"[4.*, 5.0.0)\"\n}\n```\n\n## Next Steps\n\n- Add [API Management](https://aka.ms/mcp-remote-apim-auth) to your MCP server (auth, gateway, policies, more!)\n- Add [built-in auth](https://learn.microsoft.com/en-us/azure/app-service/overview-authentication-authorization) to your MCP server\n- Enable VNET using VNET_ENABLED=true flag\n- Learn more about [related MCP efforts from Microsoft](https://github.com/microsoft/mcp/tree/main/Resources)\n\n","English","en-US",{"_27":46,"_31":50,"_36":51},{"_29":47,"_31":48,"_33":49},"Un servidor MCP personalizado basado en la nube que utiliza Azure Functions y que permite guardar y recuperar fragmentos de código con comunicación segura a través de claves, HTTPS, OAuth y opciones de aislamiento de red.","Servidor MCP remoto","# Introducción a servidores MCP remotos mediante Azure Functions (Node.js/TypeScript)\n\nEsta es una plantilla de inicio rápido para crear e implementar fácilmente un servidor MCP remoto personalizado en la nube con Azure Functions. Puede clonar, restaurar y ejecutar en su equipo local con depuración, y `azd up` para tenerlo en la nube en cuestión de minutos. El servidor MCP está protegido por diseño mediante claves y HTTPS, y permite más opciones de OAuth mediante EasyAuth o API Management, así como aislamiento de red mediante VNET.\n\n**Vea el video resumen**\n\nSi está buscando este ejemplo en más idiomas, consulte las versiones [.NET/C#](https://github.com/Azure-Samples/remote-mcp-functions-dotnet) y [Python](https://github.com/Azure-Samples/remote-mcp-functions-python) .\n\n[![Abrir en GitHub Codespaces](https://github.com/codespaces/badge.svg)](https://codespaces.new/Azure-Samples/remote-mcp-functions-typescript)\n\nA continuación se muestra el diagrama de arquitectura para el servidor MCP remoto que utiliza Azure Functions:\n\n![Diagrama de arquitectura](architecture-diagram.png)\n\n## Prerrequisitos\n\n* [Node.js](https://nodejs.org/en/download/) versión 18 o superior\n* [Herramientas principales de Azure Functions](https://learn.microsoft.com/azure/azure-functions/functions-run-local?pivots=programming-language-javascript#install-the-azure-functions-core-tools) >= `4.0.7030`\n* [CLI para desarrolladores de Azure](https://aka.ms/azd)\n* Para utilizar Visual Studio Code para ejecutar y depurar localmente:\n * [Código de Visual Studio](https://code.visualstudio.com/)\n * [Extensión de Azure Functions](https://marketplace.visualstudio.com/items?itemName=ms-azuretools.vscode-azurefunctions)\n* [Docker](https://www.docker.com/) para ejecutar Azurite, el emulador de almacenamiento de Azure (opcional)\n\n## Prepare su entorno local\n\nSe necesita un emulador de almacenamiento de Azure para este ejemplo en particular porque guardaremos y obtendremos fragmentos del almacenamiento de blobs.\n\n1. Iniciar Azurita\n\n ```shell\n docker run -p 10000:10000 -p 10001:10001 -p 10002:10002 \\\n mcr.microsoft.com/azure-storage/azurite\n ```\n\n> **Tenga en cuenta** que si usa Azurite desde la extensión VS Code, debe ejecutar `Azurite: Start` ahora o verá errores.\n\n## Ejecute su servidor MCP localmente desde la terminal\n\n1. Instalar dependencias\n\n ```shell\n npm install\n ```\n\n2. Construir el proyecto\n\n ```shell\n npm run build\n ```\n\n3. Inicie el host de Funciones localmente:\n\n ```shell\n func start\n ```\n\n> **Tenga en cuenta** que, de forma predeterminada, se usará la ruta de webhooks: `/runtime/webhooks/mcp/sse` . Más adelante, la usaremos en Azure para configurar la clave en las llamadas de cliente/host: `/runtime/webhooks/mcp/sse?code=`\n\n## Utilice el servidor MCP *local* desde un cliente/host\n\n### VS Code - Ediciones de copiloto\n\n1. **Agregue el servidor MCP** desde la paleta de comandos y agregue la URL al punto final SSE de su aplicación de función en ejecución:\n\n ```shell\n http://0.0.0.0:7071/runtime/webhooks/mcp/sse\n ```\n\n2. Seleccione **HTTP (Server-Sent-Events)** para el tipo de servidor MCP que desea agregar.\n\n3. Ingrese la URL del punto final SSE de su aplicación de función en ejecución\n\n4. Ingrese el ID del servidor. (Puede ser cualquier nombre que desee)\n\n5. Elija si desea ejecutar esto en su configuración de usuario (disponible para todas las aplicaciones) o en su configuración de espacio de trabajo (disponible solo para esta aplicación)\n\n6. **Enumere los servidores MCP** desde la paleta de comandos e inicie el servidor. Es posible que el paso anterior ya haya iniciado su servidor local. De ser así, puede omitir este paso.\n\n7. En el modo de agente de chat de Copilot, ingrese un mensaje para activar la herramienta, por ejemplo, seleccione un código e ingrese este mensaje\n\n ```plaintext\n Say Hello\n ```\n\n ```plaintext\n Save this snippet as snippet1 \n ```\n\n ```plaintext\n Retrieve snippet1 and apply to newFile.ts\n ```\n\n8. Cuando se le solicite ejecutar la herramienta, dé su consentimiento haciendo clic en **Continuar**\n\n9. Cuando haya terminado, presione Ctrl+C en la ventana de terminal para detener el proceso de host `func.exe` y **seleccione Listar servidores MCP** desde la paleta de comandos y detenga el servidor local.\n\n### Inspector de MCP\n\n1. En una **nueva ventana de terminal** , instale y ejecute MCP Inspector\n\n ```shell\n npx @modelcontextprotocol/inspector node build/index.js\n ```\n\n2. Si detuvo su aplicación de funciones anteriormente, inicie el host de funciones localmente:\n\n ```shell\n func start\n ```\n\n3. Mantenga presionada la tecla CTRL y haga clic para cargar la aplicación web MCP Inspector desde la URL que muestra la aplicación (por ejemplo, )\n\n4. Establezca el tipo de transporte en `SSE`\n\n5. Establezca la URL del punto final SSE de la aplicación de función en ejecución y **conéctese** :\n\n ```shell\n http://0.0.0.0:7071/runtime/webhooks/mcp/sse\n ```\n\n6. **Lista de herramientas** . Haga clic en una herramienta y **seleccione Ejecutar herramienta** .\n\n7. Cuando haya terminado, presione Ctrl+C en la ventana de terminal para detener el proceso de host `func.exe` , y presione Ctrl+C en la ventana de terminal para detener el proceso de host `@modelcontextprotocol/inspector` .\n\n## Implementar en Azure para MCP remoto\n\nOpcionalmente, puede optar por usar una red virtual en la muestra. (Si lo desea, hágalo antes de `azd up` ).\n\n```bash\nazd env set VNET_ENABLED true\n```\n\nEjecute este comando [azd](https://aka.ms/azd) para aprovisionar la aplicación de función, con todos los recursos de Azure necesarios, e implementar su código:\n\n```shell\nazd up\n```\n\n> **Nota:** [La administración de API]() se puede utilizar para mejorar la seguridad y las políticas de su servidor MCP, y[la autenticación integrada de App Service](https://learn.microsoft.com/en-us/azure/app-service/overview-authentication-authorization) se puede usar para configurar su proveedor OAuth favorito, incluido Entra.\n\n## Conéctese a la aplicación de función de servidor MCP *remoto* desde un cliente\n\nSu cliente necesitará una clave para invocar el nuevo punto de conexión SSE alojado, que tendrá el formato `https://.azurewebsites.net/runtime/webhooks/mcp/sse` . La función alojada requiere una clave de sistema predeterminada, que puede obtenerse desde el [portal](https://learn.microsoft.com/en-us/azure/azure-functions/function-keys-how-to?tabs=azure-portal) o la CLI ( `az functionapp keys list --resource-group --name ` ). Obtenga la clave de sistema llamada `mcp_extension` .\n\n### Conectarse al servidor MCP remoto en MCP Inspector\n\nPara MCP Inspector, puede incluir la clave en la URL:\n\n```plaintext\nhttps://.azurewebsites.net/runtime/webhooks/mcp/sse?code=\n```\n\n### Conectarse a un servidor MCP remoto en VS Code - GitHub Copilot\n\nPara GitHub Copilot en VS Code, debe establecer la clave como el encabezado `x-functions-key` en `mcp.json` y usar `https://.azurewebsites.net/runtime/webhooks/mcp/sse` como URL. El siguiente ejemplo proviene del archivo `mcp.json` incluido en este repositorio y utiliza una entrada para solicitarle que proporcione la clave al iniciar el servidor desde VS Code. Su archivo `mcp.json` tiene este aspecto:\n\n```json\n{\n \"inputs\": [\n {\n \"type\": \"promptString\",\n \"id\": \"functions-mcp-extension-system-key\",\n \"description\": \"Azure Functions MCP Extension System Key\",\n \"password\": true\n },\n {\n \"type\": \"promptString\",\n \"id\": \"functionapp-name\",\n \"description\": \"Azure Functions App Name\"\n }\n ],\n \"servers\": {\n \"remote-mcp-function\": {\n \"type\": \"sse\",\n \"url\": \"https://${input:functionapp-name}.azurewebsites.net/runtime/webhooks/mcp/sse\",\n \"headers\": {\n \"x-functions-key\": \"${input:functions-mcp-extension-system-key}\"\n }\n },\n \"local-mcp-function\": {\n \"type\": \"sse\",\n \"url\": \"http://0.0.0.0:7071/runtime/webhooks/mcp/sse\"\n }\n }\n}\n```\n\n1. Haga clic en Iniciar en el servidor `remote-mcp-function` , dentro del archivo `mcp.json` :\n\n2. Ingrese el nombre de la aplicación de función que creó en Azure Portal, cuando VS Code lo solicite.\n\n3. Introduzca la `Azure Functions MCP Extension System Key` en el mensaje. Puede copiarla desde el portal de Azure para su aplicación de funciones. Para ello, vaya al menú Funciones, seleccione Claves de la aplicación y copie la clave `mcp_extension` de Claves del sistema.\n\n4. En el modo de agente de chat de Copilot, ingrese un mensaje para activar la herramienta, por ejemplo, seleccione un código e ingrese este mensaje\n\n ```plaintext\n Say Hello\n ```\n\n ```plaintext\n Save this snippet as snippet1 \n ```\n\n ```plaintext\n Retrieve snippet1 and apply to newFile.ts\n ```\n\n## Reimplementar su código\n\nPuede ejecutar el comando `azd up` tantas veces como necesite para aprovisionar sus recursos de Azure e implementar actualizaciones de código en su aplicación de funciones.\n\n> \\[!NOTA] Los archivos de código implementados siempre se sobrescriben con el último paquete de implementación.\n\n## Limpiar recursos\n\nCuando haya terminado de trabajar con su aplicación de función y los recursos relacionados, puede usar este comando para eliminar la aplicación de función y sus recursos relacionados de Azure y evitar incurrir en costos adicionales:\n\n```shell\nazd down\n```\n\n## Código fuente\n\nEl código de función para los puntos finales `getSnippet` y `saveSnippet` se define en los archivos TypeScript del directorio `src` . Las anotaciones de las funciones MCP exponen estas funciones como herramientas del servidor MCP.\n\nEsto muestra el código para algunos ejemplos de servidor MCP (obtener cadena, obtener objeto, guardar objeto):\n\n```typescript\n// Hello function - responds with hello message\nexport async function mcpToolHello(context: InvocationContext): Promise {\n return \"Hello I am MCP Tool!\";\n}\n\n// Register the hello tool\napp.mcpTool('hello', {\n toolName: 'hello',\n description: 'Simple hello world MCP Tool that responses with a hello message.',\n handler: mcpToolHello\n});\n\n// GetSnippet function - retrieves a snippet by name\nexport async function getSnippet(_message: unknown, context: InvocationContext): Promise {\n console.info('Getting snippet');\n \n // Get snippet name from the tool arguments\n const mcptoolargs = context.triggerMetadata.mcptoolargs as { snippetname?: string };\n const snippetName = mcptoolargs?.snippetname;\n\n console.info(`Snippet name: ${snippetName}`);\n \n if (!snippetName) {\n return \"No snippet name provided\";\n }\n \n // Get the content from blob binding - properly retrieving from extraInputs\n const snippetContent = context.extraInputs.get(blobInputBinding);\n \n if (!snippetContent) {\n return `Snippet '${snippetName}' not found`;\n }\n \n console.info(`Retrieved snippet: ${snippetName}`);\n return snippetContent as string;\n}\n\n\n// Register the GetSnippet tool\napp.mcpTool('getsnippet', {\n toolName: GET_SNIPPET_TOOL_NAME,\n description: GET_SNIPPET_TOOL_DESCRIPTION,\n toolProperties: [\n {\n propertyName: SNIPPET_NAME_PROPERTY_NAME,\n propertyValue: PROPERTY_TYPE,\n description: SNIPPET_NAME_PROPERTY_DESCRIPTION,\n }\n ],\n extraInputs: [blobInputBinding],\n handler: getSnippet\n});\n\n// SaveSnippet function - saves a snippet with a name\nexport async function saveSnippet(_message: unknown, context: InvocationContext): Promise {\n console.info('Saving snippet');\n \n // Get snippet name and content from the tool arguments\n const mcptoolargs = context.triggerMetadata.mcptoolargs as { \n snippetname?: string;\n snippet?: string;\n };\n \n const snippetName = mcptoolargs?.snippetname;\n const snippet = mcptoolargs?.snippet;\n \n if (!snippetName) {\n return \"No snippet name provided\";\n }\n \n if (!snippet) {\n return \"No snippet content provided\";\n }\n \n // Save the snippet to blob storage using the output binding\n context.extraOutputs.set(blobOutputBinding, snippet);\n \n console.info(`Saved snippet: ${snippetName}`);\n return snippet;\n}\n\n// Register the SaveSnippet tool\napp.mcpTool('savesnippet', {\n toolName: SAVE_SNIPPET_TOOL_NAME,\n description: SAVE_SNIPPET_TOOL_DESCRIPTION,\n toolProperties: [\n {\n propertyName: SNIPPET_NAME_PROPERTY_NAME,\n propertyValue: PROPERTY_TYPE,\n description: SNIPPET_NAME_PROPERTY_DESCRIPTION,\n },\n {\n propertyName: SNIPPET_PROPERTY_NAME,\n propertyValue: PROPERTY_TYPE,\n description: SNIPPET_PROPERTY_DESCRIPTION,\n }\n ],\n extraOutputs: [blobOutputBinding],\n handler: saveSnippet\n});\n```\n\nTenga en cuenta que el archivo `host.json` también incluye una referencia al paquete experimental, que es necesario para las aplicaciones que usan esta función:\n\n```json\n\"extensionBundle\": {\n \"id\": \"Microsoft.Azure.Functions.ExtensionBundle.Experimental\",\n \"version\": \"[4.*, 5.0.0)\"\n}\n```\n\n## Próximos pasos\n\n* Agregue [API Management](https://aka.ms/mcp-remote-apim-auth) a su servidor MCP (autorización, puerta de enlace, políticas, ¡y más!)\n* Agregue[autenticación integrada](https://learn.microsoft.com/en-us/azure/app-service/overview-authentication-authorization) a su servidor MCP\n* Habilitar VNET usando el indicador VNET\\_ENABLED=true\n* Obtenga más información sobre [los esfuerzos relacionados con MCP de Microsoft](https://github.com/microsoft/mcp/tree/main/Resources)","Spanish","es-ES",{"_27":53,"_31":57,"_36":58},{"_29":54,"_31":55,"_33":56},"Azure Functions を使用したクラウドベースのカスタム MCP サーバー。キー、HTTPS、OAuth、ネットワーク分離オプションを介した安全な通信でコードスニペットを保存および取得できます。","リモートMCPサーバー","# Azure Functions を使用したリモート MCP サーバーの使用開始 (Node.js/TypeScript)\n\nこれは、Azure Functions を使用してカスタムリモート MCP サーバーを簡単に構築し、クラウドにデプロイするためのクイックスタートテンプレートです。ローカルマシンでクローン作成、復元、実行（デバッグ付き）し、数分でクラウドに`azd up`できます。MCP サーバーは、キーと HTTPS を使用して設計上セキュリティ保護されており、EasyAuth や API Management を使用した OAuth のオプションや、VNET を使用したネットワーク分離もサポートします。\n\n**ビデオの概要を見る**\n\nこのサンプルを他の言語で探している場合は[、.NET/C#](https://github.com/Azure-Samples/remote-mcp-functions-dotnet)および[Python](https://github.com/Azure-Samples/remote-mcp-functions-python)バージョンを確認してください。\n\n[![GitHubコードスペースで開く](https://github.com/codespaces/badge.svg)](https://codespaces.new/Azure-Samples/remote-mcp-functions-typescript)\n\n以下は、Azure Functions を使用したリモート MCP サーバーのアーキテクチャ図です。\n\n![アーキテクチャ図](architecture-diagram.png)\n\n## 前提条件\n\n* [Node.js](https://nodejs.org/en/download/)バージョン 18 以上\n* [Azure Functions コアツール](https://learn.microsoft.com/azure/azure-functions/functions-run-local?pivots=programming-language-javascript#install-the-azure-functions-core-tools)>= `4.0.7030`\n* [Azure 開発者 CLI](https://aka.ms/azd)\n* Visual Studio Code を使用してローカルで実行およびデバッグするには:\n * [ビジュアルスタジオコード](https://code.visualstudio.com/)\n * [Azure Functions 拡張機能](https://marketplace.visualstudio.com/items?itemName=ms-azuretools.vscode-azurefunctions)\n* Azure ストレージエミュレーターである Azurite を実行するための[Docker](https://www.docker.com/) (オプション)\n\n## ローカル環境を準備する\n\nこの特定のサンプルでは、BLOB ストレージからスニペットを保存および取得するため、Azure ストレージエミュレーターが必要です。\n\n1. アズライトを始める\n\n ```shell\n docker run -p 10000:10000 -p 10001:10001 -p 10002:10002 \\\n mcr.microsoft.com/azure-storage/azurite\n ```\n\n> **注意:** VS Code 拡張機能から取得した Azurite を使用する場合は`Azurite: Start` now」を実行する必要があります。そうしないとエラーが表示されます。\n\n## ターミナルからMCPサーバーをローカルで実行する\n\n1. 依存関係をインストールする\n\n ```shell\n npm install\n ```\n\n2. プロジェクトを構築する\n\n ```shell\n npm run build\n ```\n\n3. Functions ホストをローカルで起動します。\n\n ```shell\n func start\n ```\n\n> **デフォルト**では、Webhooksルート（ `/runtime/webhooks/mcp/sse`が使用されます。後ほど、Azureでクライアント/ホスト呼び出し時にキーを設定するために、 `/runtime/webhooks/mcp/sse?code=`\n\n## クライアント/ホスト内から*ローカル*MCP サーバーを使用する\n\n### VS Code - コパイロット編集\n\n1. コマンドパレットから**MCP サーバーを追加し**、実行中の Function App の SSE エンドポイントに URL を追加します。\n\n ```shell\n http://0.0.0.0:7071/runtime/webhooks/mcp/sse\n ```\n\n2. 追加する MCP サーバーのタイプとして\\*\\*HTTP (Server-Sent-Events)\\*\\*を選択します。\n\n3. 実行中の関数アプリのSSEエンドポイントのURLを入力します\n\n4. サーバーIDを入力します。（任意の名前を指定できます）\n\n5. これをユーザー設定（すべてのアプリで利用可能）で実行するか、ワークスペース設定（このアプリでのみ利用可能）で実行するかを選択します。\n\n6. コマンドパレットから**MCPサーバーの一覧を表示し**、サーバーを起動します。前の手順でローカルサーバーが既に起動している可能性があります。その場合は、この手順をスキップしてください。\n\n7. Copilot チャットエージェントモードでは、ツールを起動するためのプロンプトを入力します。たとえば、コードを選択してこのプロンプトを入力します。\n\n ```plaintext\n Say Hello\n ```\n\n ```plaintext\n Save this snippet as snippet1 \n ```\n\n ```plaintext\n Retrieve snippet1 and apply to newFile.ts\n ```\n\n8. ツールを実行するように求められたら、 **「続行」を**クリックして同意します。\n\n9. 完了したら、ターミナルウィンドウで Ctrl + C を押して`func.exe`ホストプロセスを停止し、コマンドパレットから**MCP サーバーを一覧表示して**ローカルサーバーを停止します。\n\n### MCP検査官\n\n1. **新しいターミナルウィンドウ**で、MCP Inspectorをインストールして実行します。\n\n ```shell\n npx @modelcontextprotocol/inspector node build/index.js\n ```\n\n2. 以前に関数アプリを停止した場合は、Functions ホストをローカルで起動します。\n\n ```shell\n func start\n ```\n\n3. Ctrl キーを押しながらクリックすると、アプリによって表示される URL (例: ) から MCP Inspector Web アプリが読み込まれます。\n\n4. トランスポートタイプを`SSE`に設定する\n\n5. 実行中の Function App の SSE エンドポイントへの URL を設定し、**接続します**。\n\n ```shell\n http://0.0.0.0:7071/runtime/webhooks/mcp/sse\n ```\n\n6. **ツールの一覧を表示します**。ツールをクリックして**ツールを実行してください**。\n\n7. 完了したら、ターミナルウィンドウで Ctrl + C キーを押して`func.exe`ホストプロセスを停止し、ターミナルウィンドウで Ctrl + C キーを押して`@modelcontextprotocol/inspector`ホストプロセスを停止します。\n\n## リモート MCP 用に Azure にデプロイする\n\nオプションで、サンプルで使用されている VNet をオプトインできます。(これを選択する場合は、 `azd up`の前に実行してください)\n\n```bash\nazd env set VNET_ENABLED true\n```\n\n次の[azd](https://aka.ms/azd)コマンドを実行して、必要な Azure リソースを使用して関数アプリをプロビジョニングし、コードをデプロイします。\n\n```shell\nazd up\n```\n\n> **注:** [API Management を]()使用すると、MCP サーバーのセキュリティとポリシーを強化できます。また、[App Service の組み込み認証を](https://learn.microsoft.com/en-us/azure/app-service/overview-authentication-authorization)使用すると、Entra などのお気に入りの OAuth プロバイダーを設定できます。\n\n## クライアントから*リモート*MCP サーバー関数アプリに接続する\n\nクライアントは、新しいホスト型SSEエンドポイントを呼び出すためにキーを必要とします。このキー`https://.azurewebsites.net/runtime/webhooks/mcp/sse`という形式になります。ホスト型関数には、デフォルトでシステムキーが必要です。これは[ポータル](https://learn.microsoft.com/en-us/azure/azure-functions/function-keys-how-to?tabs=azure-portal)またはCLI（ `az functionapp keys list --resource-group --name ` ）から取得できます`mcp_extension`という名前のシステムキーを取得してください。\n\n### MCP InspectorでリモートMCPサーバーに接続する\n\nMCP Inspector の場合、URL にキーを含めることができます。\n\n```plaintext\nhttps://.azurewebsites.net/runtime/webhooks/mcp/sse?code=\n```\n\n### VS CodeでリモートMCPサーバーに接続する - GitHub Copilot\n\nVS Code内でGitHub Copilotを使用する場合は、 `mcp.json`の`x-functions-key`ヘッダーにキーを設定し、URLには`https://.azurewebsites.net/runtime/webhooks/mcp/sse`を使用します。以下の例は、このリポジトリに含まれる`mcp.json`ファイ��からの抜粋です。VS Codeからサーバーを起動する際に、入力欄を使用してキーの入力を求めます`mcp.json`ファイルは以下のようになります。\n\n```json\n{\n \"inputs\": [\n {\n \"type\": \"promptString\",\n \"id\": \"functions-mcp-extension-system-key\",\n \"description\": \"Azure Functions MCP Extension System Key\",\n \"password\": true\n },\n {\n \"type\": \"promptString\",\n \"id\": \"functionapp-name\",\n \"description\": \"Azure Functions App Name\"\n }\n ],\n \"servers\": {\n \"remote-mcp-function\": {\n \"type\": \"sse\",\n \"url\": \"https://${input:functionapp-name}.azurewebsites.net/runtime/webhooks/mcp/sse\",\n \"headers\": {\n \"x-functions-key\": \"${input:functions-mcp-extension-system-key}\"\n }\n },\n \"local-mcp-function\": {\n \"type\": \"sse\",\n \"url\": \"http://0.0.0.0:7071/runtime/webhooks/mcp/sse\"\n }\n }\n}\n```\n\n1. `mcp.json`ファイル内のサーバー`remote-mcp-function`で \\[開始] をクリックします。\n\n2. VS Code からプロンプトが表示されたら、Azure Portal で作成した関数アプリの名前を入力します。\n\n3. プロンプトに`Azure Functions MCP Extension System Key`を入力します。Azureポータルから関数アプリの「Functions」メニュー項目に移動し、「アプリキー」を選択して、「システムキー」から`mcp_extension`キーをコピーすることで、このキーをコピーできます。\n\n4. Copilot チャットエージェントモードでは、ツールを起動するためのプロンプトを入力します。たとえば、コードを選択してこのプロンプトを入力します。\n\n ```plaintext\n Say Hello\n ```\n\n ```plaintext\n Save this snippet as snippet1 \n ```\n\n ```plaintext\n Retrieve snippet1 and apply to newFile.ts\n ```\n\n## コードを再デプロイする\n\nAzure リソースをプロビジョニングし、関数アプリにコードの更新をデプロイするために必要な回数だけ`azd up`コマンドを実行できます。\n\n> \\[!NOTE] デプロイされたコードファイルは常に最新のデプロイパッケージによって上書きされます。\n\n## リソースをクリーンアップする\n\n関数アプリと関連リソースの操作が完了したら、次のコマンドを使用して、関数アプリと関連リソースを Azure から削除し、それ以上のコストが発生しないようにすることができます。\n\n```shell\nazd down\n```\n\n## ソースコード\n\n`getSnippet`および`saveSnippet`エンドポイントの関数コードは、 `src`ディレクトリ内の TypeScript ファイルで定義されています。MCP 関数アノテーションは、これらの関数を MCP サーバーツールとして公開します。\n\nこれは、いくつかの MCP サーバー例 (文字列の取得、オブジェクトの取得、オブジェクトの保存) のコードを示しています。\n\n```typescript\n// Hello function - responds with hello message\nexport async function mcpToolHello(context: InvocationContext): Promise {\n return \"Hello I am MCP Tool!\";\n}\n\n// Register the hello tool\napp.mcpTool('hello', {\n toolName: 'hello',\n description: 'Simple hello world MCP Tool that responses with a hello message.',\n handler: mcpToolHello\n});\n\n// GetSnippet function - retrieves a snippet by name\nexport async function getSnippet(_message: unknown, context: InvocationContext): Promise {\n console.info('Getting snippet');\n \n // Get snippet name from the tool arguments\n const mcptoolargs = context.triggerMetadata.mcptoolargs as { snippetname?: string };\n const snippetName = mcptoolargs?.snippetname;\n\n console.info(`Snippet name: ${snippetName}`);\n \n if (!snippetName) {\n return \"No snippet name provided\";\n }\n \n // Get the content from blob binding - properly retrieving from extraInputs\n const snippetContent = context.extraInputs.get(blobInputBinding);\n \n if (!snippetContent) {\n return `Snippet '${snippetName}' not found`;\n }\n \n console.info(`Retrieved snippet: ${snippetName}`);\n return snippetContent as string;\n}\n\n\n// Register the GetSnippet tool\napp.mcpTool('getsnippet', {\n toolName: GET_SNIPPET_TOOL_NAME,\n description: GET_SNIPPET_TOOL_DESCRIPTION,\n toolProperties: [\n {\n propertyName: SNIPPET_NAME_PROPERTY_NAME,\n propertyValue: PROPERTY_TYPE,\n description: SNIPPET_NAME_PROPERTY_DESCRIPTION,\n }\n ],\n extraInputs: [blobInputBinding],\n handler: getSnippet\n});\n\n// SaveSnippet function - saves a snippet with a name\nexport async function saveSnippet(_message: unknown, context: InvocationContext): Promise {\n console.info('Saving snippet');\n \n // Get snippet name and content from the tool arguments\n const mcptoolargs = context.triggerMetadata.mcptoolargs as { \n snippetname?: string;\n snippet?: string;\n };\n \n const snippetName = mcptoolargs?.snippetname;\n const snippet = mcptoolargs?.snippet;\n \n if (!snippetName) {\n return \"No snippet name provided\";\n }\n \n if (!snippet) {\n return \"No snippet content provided\";\n }\n \n // Save the snippet to blob storage using the output binding\n context.extraOutputs.set(blobOutputBinding, snippet);\n \n console.info(`Saved snippet: ${snippetName}`);\n return snippet;\n}\n\n// Register the SaveSnippet tool\napp.mcpTool('savesnippet', {\n toolName: SAVE_SNIPPET_TOOL_NAME,\n description: SAVE_SNIPPET_TOOL_DESCRIPTION,\n toolProperties: [\n {\n propertyName: SNIPPET_NAME_PROPERTY_NAME,\n propertyValue: PROPERTY_TYPE,\n description: SNIPPET_NAME_PROPERTY_DESCRIPTION,\n },\n {\n propertyName: SNIPPET_PROPERTY_NAME,\n propertyValue: PROPERTY_TYPE,\n description: SNIPPET_PROPERTY_DESCRIPTION,\n }\n ],\n extraOutputs: [blobOutputBinding],\n handler: saveSnippet\n});\n```\n\n`host.json`ファイルには、この機能を利用するアプリに必要な実験的なバンドルへの参照も含まれていることに注意してください。\n\n```json\n\"extensionBundle\": {\n \"id\": \"Microsoft.Azure.Functions.ExtensionBundle.Experimental\",\n \"version\": \"[4.*, 5.0.0)\"\n}\n```\n\n## 次のステップ\n\n* MCP サーバーに[API 管理](https://aka.ms/mcp-remote-apim-auth)を追加します (認証、ゲートウェイ、ポリシーなど)\n* MCPサーバーに[組み込み認証](https://learn.microsoft.com/en-us/azure/app-service/overview-authentication-authorization)を追加する\n* VNET\\_ENABLED=true フラグを使用して VNET を有効にする\n* [Microsoft の MCP 関連取り組み](https://github.com/microsoft/mcp/tree/main/Resources)について詳しくは、こちらをご覧ください。","Japanese","ja-JP",{"_27":60,"_31":64,"_36":65},{"_29":61,"_31":62,"_33":63},"키, HTTPS, OAuth 및 네트워크 격리 옵션을 통한 보안 통신으로 코드 조각을 저장하고 검색할 수 있는 Azure Functions를 사용하는 클라우드 기반 사용자 지정 MCP 서버입니다.","원격 MCP 서버","# Azure Functions를 사용하여 원격 MCP 서버 시작하기(Node.js/TypeScript)\n\nAzure Functions를 사용하여 사용자 지정 원격 MCP 서버를 클라우드에 쉽게 구축하고 배포할 수 있는 빠른 시작 템플릿입니다. 디버깅을 통해 로컬 컴퓨터에서 복제/복원/실행하고, 몇 분 안에 클라우드에 배포 `azd up` . MCP 서버는 키와 HTTP를 사용하여 보안이 강화되었으며, EasyAuth 및/또는 API Management를 사용한 OAuth 및 VNET을 사용한 네트워크 격리 등 다양한 옵션을 제공합니다.\n\n**비디오 개요를 시청하세요**\n\n더 많은 언어로 이 샘플을 찾고 있다면 [.NET/C#](https://github.com/Azure-Samples/remote-mcp-functions-dotnet) 및 [Python](https://github.com/Azure-Samples/remote-mcp-functions-python) 버전을 확인해보세요.\n\n[![GitHub Codespaces에서 열기](https://github.com/codespaces/badge.svg)](https://codespaces.new/Azure-Samples/remote-mcp-functions-typescript)\n\nAzure Functions를 사용하는 원격 MCP 서버에 대한 아키텍처 다이어그램은 다음과 같습니다.\n\n![아키텍처 다이어그램](architecture-diagram.png)\n\n## 필수 조건\n\n* [Node.js](https://nodejs.org/en/download/) 버전 18 이상\n* [Azure Functions 핵심 도구](https://learn.microsoft.com/azure/azure-functions/functions-run-local?pivots=programming-language-javascript#install-the-azure-functions-core-tools) >= `4.0.7030`\n* [Azure 개발자 CLI](https://aka.ms/azd)\n* Visual Studio Code를 사용하여 로컬에서 실행하고 디버깅하려면:\n * [비주얼 스튜디오 코드](https://code.visualstudio.com/)\n * [Azure Functions 확장](https://marketplace.visualstudio.com/items?itemName=ms-azuretools.vscode-azurefunctions)\n* Azure Storage 에뮬레이터인 Azurite를 실행하기 위한 [Docker](https://www.docker.com/) (선택 사항)\n\n## 지역 환경을 준비하세요\n\n이 특정 샘플에는 Blob 저장소에서 스니펫을 저장하고 가져오기 때문에 Azure Storage Emulator가 필요합니다.\n\n1. Azurite 시작하기\n\n 지엑스피1\n\n> VS Code 확장 프로그램에서 Azurite를 사용하는 **경우** `Azurite: Start` 하지 않으면 오류가 발생합니다.\n\n## 터미널에서 로컬로 MCP 서버를 실행하세요.\n\n1. 종속성 설치\n\n ```shell\n npm install\n ```\n\n2. 프로젝트를 빌드하세요\n\n ```shell\n npm run build\n ```\n\n3. 로컬에서 Functions 호스트를 시작합니다.\n\n ```shell\n func start\n ```\n\n> **기본적** 으로 웹훅 경로( `/runtime/webhooks/mcp/sse` 사용합니다. 나중에 Azure에서 이 경로를 사용하여 클라이언트/호스트 호출 시 키를 설정합니다( `/runtime/webhooks/mcp/sse?code=`\n\n## 클라이언트/호스트 내에서 *로컬* MCP 서버를 사용합니다.\n\n### VS 코드 - Copilot 편집\n\n1. 명령 팔레트에서 **MCP 서버를 추가하고** 실행 중인 함수 앱의 SSE 엔드포인트에 URL을 추가합니다.\n\n ```shell\n http://0.0.0.0:7071/runtime/webhooks/mcp/sse\n ```\n\n2. 추가할 MCP 서버 유형으로 **HTTP(서버에서 보낸 이벤트)를** 선택합니다.\n\n3. 실행 중인 함수 앱의 SSE 엔드포인트에 대한 URL을 입력하세요.\n\n4. 서버 ID를 입력하세요. (원하는 이름을 입력하세요)\n\n5. 사용자 설정(모든 앱에서 사용 가능)에서 실행할지, 작업 공간 설정(이 앱에서만 사용 가능)에서 실행할지 선택하세요.\n\n6. 명령 팔레트에서 **MCP 서버를 나열** 하고 서버를 시작하세요. 이전 단계에서 로컬 서버가 이미 시작되었을 수 있습니다. 그렇다면 이 단계를 건너뛸 수 있습니다.\n\n7. Copilot 채팅 에이전트 모드에서 도구를 트리거하기 위한 프롬프트를 입력합니다. 예를 들어, 일부 코드를 선택하고 이 프롬프트를 입력합니다.\n\n ```plaintext\n Say Hello\n ```\n\n ```plaintext\n Save this snippet as snippet1 \n ```\n\n ```plaintext\n Retrieve snippet1 and apply to newFile.ts\n ```\n\n8. 도구를 실행하라는 메시지가 표시되면 **계속을** 클릭하여 동의합니다.\n\n9. 완료되면 터미널 창에서 Ctrl+C를 눌러 `func.exe` 호스트 프로세스를 중지하고 명령 팔레트에서 **MCP 서버 목록을 표시하여** 로컬 서버를 중지합니다.\n\n### MCP 검사관\n\n1. **새 터미널 창** 에서 MCP Inspector를 설치하고 실행하세요.\n\n ```shell\n npx @modelcontextprotocol/inspector node build/index.js\n ```\n\n2. 이전에 함수 앱을 중지한 경우 로컬에서 함수 호스트를 시작합니다.\n\n ```shell\n func start\n ```\n\n3. CTRL을 클릭하면 앱에 표시된 URL(예: )에서 MCP Inspector 웹 앱을 로드합니다.\n\n4. 전송 유형을 `SSE` 로 설정하세요\n\n5. 실행 중인 Function 앱의 SSE 엔드포인트에 URL을 설정하고 **Connect를 실행합니다** .\n\n ```shell\n http://0.0.0.0:7071/runtime/webhooks/mcp/sse\n ```\n\n6. **도구 목록** . 도구를 클릭하고 **도구 실행** .\n\n7. 완료되면 터미널 창에서 Ctrl+C를 눌러 `func.exe` 호스트 프로세스를 중지하고, 터미널 창에서 Ctrl+C를 눌러 `@modelcontextprotocol/inspector` 호스트 프로세스를 중지합니다.\n\n## 원격 MCP를 위해 Azure에 배포\n\n선택적으로 샘플에서 사용되는 VNet에 옵트인할 수 있습니다. (선택하는 경우 `azd up` 전에 옵트인하세요.)\n\n```bash\nazd env set VNET_ENABLED true\n```\n\n이 [azd](https://aka.ms/azd) 명령을 실행하여 필요한 Azure 리소스로 함수 앱을 프로비저닝하고 코드를 배포합니다.\n\n```shell\nazd up\n```\n\n> [API 관리를]() **사용** 하면 MCP 서버의 보안과 정책을 개선할 수 있으며[App Service 내장 인증을](https://learn.microsoft.com/en-us/azure/app-service/overview-authentication-authorization) 사용하여 Entra를 포함한 원하는 OAuth 공급자를 설정할 수 있습니다.\n\n## 클라이언트에서 *원격* MCP 서버 기능 앱에 연결\n\n클라이언트는 새로 호스팅된 SSE 엔드포인트를 호출하기 위해 키가 필요합니다. 키는 `https://.azurewebsites.net/runtime/webhooks/mcp/sse` 형식입니다. 호스팅된 함수에는 기본적으로 시스템 키가 필요하며, [포털](https://learn.microsoft.com/en-us/azure/azure-functions/function-keys-how-to?tabs=azure-portal) 이나 CLI( `az functionapp keys list --resource-group --name ` )에서 얻을 수 있습니다. `mcp_extension` 이라는 이름의 시스템 키를 가져오세요.\n\n### MCP Inspector에서 원격 MCP 서버에 연결\n\nMCP Inspector의 경우 URL에 키를 포함할 수 있습니다.\n\n```plaintext\nhttps://.azurewebsites.net/runtime/webhooks/mcp/sse?code=\n```\n\n### VS Code에서 원격 MCP 서버에 연결 - GitHub Copilot\n\nVS Code 내 GitHub Copilot의 경우, `mcp.json` 의 `x-functions-key` 헤더에 키를 설정하고 URL에는 `https://.azurewebsites.net/runtime/webhooks/mcp/sse` 사용해야 합니다. 다음 예시는 이 저장소에 포함된 `mcp.json` 파일에서 가져온 것으로, VS Code에서 서버를 시작할 때 입력을 통해 키를 입력하라는 메시지를 표시합니다. `mcp.json` 파일은 다음과 같습니다.\n\n```json\n{\n \"inputs\": [\n {\n \"type\": \"promptString\",\n \"id\": \"functions-mcp-extension-system-key\",\n \"description\": \"Azure Functions MCP Extension System Key\",\n \"password\": true\n },\n {\n \"type\": \"promptString\",\n \"id\": \"functionapp-name\",\n \"description\": \"Azure Functions App Name\"\n }\n ],\n \"servers\": {\n \"remote-mcp-function\": {\n \"type\": \"sse\",\n \"url\": \"https://${input:functionapp-name}.azurewebsites.net/runtime/webhooks/mcp/sse\",\n \"headers\": {\n \"x-functions-key\": \"${input:functions-mcp-extension-system-key}\"\n }\n },\n \"local-mcp-function\": {\n \"type\": \"sse\",\n \"url\": \"http://0.0.0.0:7071/runtime/webhooks/mcp/sse\"\n }\n }\n}\n```\n\n1. `mcp.json` 파일 내부의 `remote-mcp-function` 서버에서 시작을 클릭합니다.\n\n2. VS Code에서 메시지가 표시되면 Azure Portal에서 만든 함수 앱의 이름을 입력합니다.\n\n3. 프롬프트에 `Azure Functions MCP Extension System Key` 입력하세요. 함수 앱의 Azure Portal에서 이 키를 복사하려면 함수 메뉴 항목 > 앱 키로 이동한 후 시스템 키에서 `mcp_extension` 키를 복사하세요.\n\n4. Copilot 채팅 에이전트 모드에서 도구를 트리거하기 위한 프롬프트를 입력합니다. 예를 들어, 일부 코드를 선택하고 이 프롬프트를 입력합니다.\n\n ```plaintext\n Say Hello\n ```\n\n ```plaintext\n Save this snippet as snippet1 \n ```\n\n ```plaintext\n Retrieve snippet1 and apply to newFile.ts\n ```\n\n## 코드를 다시 배포하세요\n\nAzure 리소스를 프로비저닝하고 함수 앱에 코드 업데이트를 배포하기 위해 필요한 만큼 `azd up` 명령을 실행할 수 있습니다.\n\n> \\[!NOTE] 배포된 코드 파일은 항상 최신 배포 패키지로 덮어쓰여집니다.\n\n## 자원 정리\n\n함수 앱과 관련 리소스 작업이 끝나면 이 명령을 사용하여 Azure에서 함수 앱과 관련 리소스를 삭제하고 추가 비용이 발생하지 않도록 할 수 있습니다.\n\n```shell\nazd down\n```\n\n## 소스 코드\n\n`getSnippet` 및 `saveSnippet` 엔드포인트에 대한 함수 코드는 `src` 디렉터리의 TypeScript 파일에 정의되어 있습니다. MCP 함수 애노테이션은 이러한 함수를 MCP 서버 도구로 제공합니다.\n\n다음은 몇 가지 MCP 서버 예제(문자열 가져오기, 객체 가져오기, 객체 저장)에 대한 코드입니다.\n\n```typescript\n// Hello function - responds with hello message\nexport async function mcpToolHello(context: InvocationContext): Promise {\n return \"Hello I am MCP Tool!\";\n}\n\n// Register the hello tool\napp.mcpTool('hello', {\n toolName: 'hello',\n description: 'Simple hello world MCP Tool that responses with a hello message.',\n handler: mcpToolHello\n});\n\n// GetSnippet function - retrieves a snippet by name\nexport async function getSnippet(_message: unknown, context: InvocationContext): Promise {\n console.info('Getting snippet');\n \n // Get snippet name from the tool arguments\n const mcptoolargs = context.triggerMetadata.mcptoolargs as { snippetname?: string };\n const snippetName = mcptoolargs?.snippetname;\n\n console.info(`Snippet name: ${snippetName}`);\n \n if (!snippetName) {\n return \"No snippet name provided\";\n }\n \n // Get the content from blob binding - properly retrieving from extraInputs\n const snippetContent = context.extraInputs.get(blobInputBinding);\n \n if (!snippetContent) {\n return `Snippet '${snippetName}' not found`;\n }\n \n console.info(`Retrieved snippet: ${snippetName}`);\n return snippetContent as string;\n}\n\n\n// Register the GetSnippet tool\napp.mcpTool('getsnippet', {\n toolName: GET_SNIPPET_TOOL_NAME,\n description: GET_SNIPPET_TOOL_DESCRIPTION,\n toolProperties: [\n {\n propertyName: SNIPPET_NAME_PROPERTY_NAME,\n propertyValue: PROPERTY_TYPE,\n description: SNIPPET_NAME_PROPERTY_DESCRIPTION,\n }\n ],\n extraInputs: [blobInputBinding],\n handler: getSnippet\n});\n\n// SaveSnippet function - saves a snippet with a name\nexport async function saveSnippet(_message: unknown, context: InvocationContext): Promise {\n console.info('Saving snippet');\n \n // Get snippet name and content from the tool arguments\n const mcptoolargs = context.triggerMetadata.mcptoolargs as { \n snippetname?: string;\n snippet?: string;\n };\n \n const snippetName = mcptoolargs?.snippetname;\n const snippet = mcptoolargs?.snippet;\n \n if (!snippetName) {\n return \"No snippet name provided\";\n }\n \n if (!snippet) {\n return \"No snippet content provided\";\n }\n \n // Save the snippet to blob storage using the output binding\n context.extraOutputs.set(blobOutputBinding, snippet);\n \n console.info(`Saved snippet: ${snippetName}`);\n return snippet;\n}\n\n// Register the SaveSnippet tool\napp.mcpTool('savesnippet', {\n toolName: SAVE_SNIPPET_TOOL_NAME,\n description: SAVE_SNIPPET_TOOL_DESCRIPTION,\n toolProperties: [\n {\n propertyName: SNIPPET_NAME_PROPERTY_NAME,\n propertyValue: PROPERTY_TYPE,\n description: SNIPPET_NAME_PROPERTY_DESCRIPTION,\n },\n {\n propertyName: SNIPPET_PROPERTY_NAME,\n propertyValue: PROPERTY_TYPE,\n description: SNIPPET_PROPERTY_DESCRIPTION,\n }\n ],\n extraOutputs: [blobOutputBinding],\n handler: saveSnippet\n});\n```\n\n`host.json` 파일에는 이 기능을 사용하는 앱에 필요한 실험 번들에 대한 참조도 포함되어 있습니다.\n\n```json\n\"extensionBundle\": {\n \"id\": \"Microsoft.Azure.Functions.ExtensionBundle.Experimental\",\n \"version\": \"[4.*, 5.0.0)\"\n}\n```\n\n## 다음 단계\n\n* MCP 서버에 [API 관리](https://aka.ms/mcp-remote-apim-auth) (인증, 게이트웨이, 정책 등!)를 추가하세요.\n* MCP 서버에[내장 인증](https://learn.microsoft.com/en-us/azure/app-service/overview-authentication-authorization) 추가\n* VNET\\_ENABLED=true 플래그를 사용하여 VNET을 활성화합니다.\n* [Microsoft의 관련 MCP 활동](https://github.com/microsoft/mcp/tree/main/Resources) 에 대해 자세히 알아보세요.","Korean","ko-KR",{"_27":67,"_31":71,"_36":72},{"_29":68,"_31":69,"_33":70},"Облачный настраиваемый сервер MCP, использующий Azure Functions, который позволяет сохранять и извлекать фрагменты кода с защищенной связью через ключи, HTTPS, OAuth и параметры сетевой изоляции.","Удаленный MCP-сервер","# Начало работы с удаленными серверами MCP с использованием функций Azure (Node.js/TypeScript)\n\nЭто шаблон быстрого старта для простого создания и развертывания настраиваемого удаленного сервера MCP в облаке с использованием функций Azure. Вы можете клонировать/восстановить/запустить на локальной машине с отладкой и `azd up` , чтобы разместить его в облаке за пару минут. Сервер MCP защищен по умолчанию с помощью ключей и HTTP и предоставляет больше возможностей для OAuth с использованием EasyAuth и/или API Management, а также сетевой изоляции с использованием VNET.\n\n**Посмотрите видеообзор**\n\nЕсли вы ищете этот пример на других языках, ознакомьтесь с версиями [.NET/C#](https://github.com/Azure-Samples/remote-mcp-functions-dotnet) и [Python](https://github.com/Azure-Samples/remote-mcp-functions-python) .\n\n[![Открыть в GitHub Codespaces](https://github.com/codespaces/badge.svg)](https://codespaces.new/Azure-Samples/remote-mcp-functions-typescript)\n\nНиже представлена схема архитектуры удаленного сервера MCP с использованием функций Azure:\n\n![Архитектурная схема](architecture-diagram.png)\n\n## Предпосылки\n\n* [Node.js](https://nodejs.org/en/download/) версии 18 или выше\n* [Основные инструменты Azure Functions](https://learn.microsoft.com/azure/azure-functions/functions-run-local?pivots=programming-language-javascript#install-the-azure-functions-core-tools) >= `4.0.7030`\n* [Интерфейс командной строки разработчика Azure](https://aka.ms/azd)\n* Чтобы использовать Visual Studio Code для локального запуска и отладки:\n * [Код Visual Studio](https://code.visualstudio.com/)\n * [Расширение функций Azure](https://marketplace.visualstudio.com/items?itemName=ms-azuretools.vscode-azurefunctions)\n* [Docker](https://www.docker.com/) для запуска Azurite, эмулятора хранилища Azure (необязательно)\n\n## Подготовьте местную среду\n\nДля этого конкретного примера необходим эмулятор хранилища Azure, поскольку мы будем сохранять и получать фрагменты из хранилища BLOB-объектов.\n\n1. Старт Азурита\n\n ```shell\n docker run -p 10000:10000 -p 10001:10001 -p 10002:10002 \\\n mcr.microsoft.com/azure-storage/azurite\n ```\n\n> **Обратите внимание,** если вы используете Azurite из расширения VS Code, вам необходимо запустить `Azurite: Start` now, иначе вы увидите ошибки.\n\n## Запустите свой MCP-сервер локально из терминала\n\n1. Установить зависимости\n\n ```shell\n npm install\n ```\n\n2. Построить проект\n\n ```shell\n npm run build\n ```\n\n3. Запустите хост функций локально:\n\n ```shell\n func start\n ```\n\n> **Обратите внимание,** что по умолчанию это будет использовать маршрут webhooks: `/runtime/webhooks/mcp/sse` . Позже мы будем использовать это в Azure для установки ключа для вызовов клиент/хост: `/runtime/webhooks/mcp/sse?code=`\n\n## Используйте *локальный* сервер MCP из клиента/хоста\n\n### VS Code - Правки второго пилота\n\n1. **Добавьте MCP-сервер** из палитры команд и добавьте URL-адрес к конечной точке SSE вашего работающего приложения-функции:\n\n ```shell\n http://0.0.0.0:7071/runtime/webhooks/mcp/sse\n ```\n\n2. Выберите **HTTP (Server-Sent-Events)** в качестве типа добавляемого сервера MCP.\n\n3. Введите URL-адрес конечной точки SSE вашего работающего приложения-функции.\n\n4. Введите идентификатор сервера. (Это может быть любое имя по вашему желанию)\n\n5. Выберите, хотите ли вы запустить это в настройках пользователя (доступно для всех ваших приложений) или в настройках рабочей области (доступно только для этого приложения)\n\n6. **Выведите список серверов MCP** из палитры команд и запустите сервер. Предыдущий шаг мог уже запустить ваш локальный сервер. Если так, вы можете пропустить этот шаг.\n\n7. В режиме чат-агента Copilot введите запрос для запуска инструмента, например, выберите какой-либо код и введите этот запрос.\n\n ```plaintext\n Say Hello\n ```\n\n ```plaintext\n Save this snippet as snippet1 \n ```\n\n ```plaintext\n Retrieve snippet1 and apply to newFile.ts\n ```\n\n8. При появлении запроса на запуск инструмента дайте согласие, нажав **«Продолжить».**\n\n9. Закончив, нажмите Ctrl+C в окне терминала, чтобы остановить хост-процесс `func.exe` , а затем **выведите список серверов MCP** из палитры команд и остановите локальный сервер.\n\n### Инспектор МКП\n\n1. В **новом окне терминала** установите и запустите MCP Inspector.\n\n ```shell\n npx @modelcontextprotocol/inspector node build/index.js\n ```\n\n2. Если вы ранее остановили приложение-функцию, запустите хост-функцию локально:\n\n ```shell\n func start\n ```\n\n3. Удерживая клавишу CTRL, щелкните, чтобы загрузить веб-приложение MCP Inspector с URL-адреса, отображаемого приложением (например, ).\n\n4. Установите тип транспорта на `SSE`\n\n5. Укажите URL-адрес конечной точки SSE вашего работающего приложения-функции и **подключитесь** :\n\n ```shell\n http://0.0.0.0:7071/runtime/webhooks/mcp/sse\n ```\n\n6. **Список инструментов** . Щелкните по инструменту и **запустите инструмент** .\n\n7. Когда вы закончите, нажмите Ctrl+C в окне терминала, чтобы остановить хост-процесс `func.exe` , и нажмите Ctrl+C в окне терминала, чтобы остановить хост-процесс `@modelcontextprotocol/inspector` .\n\n## Развертывание в Azure для удаленного MCP\n\nПри желании вы можете согласиться на использование VNet в образце. (Если вы выберете этот вариант, сделайте это до `azd up` )\n\n```bash\nazd env set VNET_ENABLED true\n```\n\nЗапустите эту команду [azd](https://aka.ms/azd) , чтобы предоставить приложению-функции все необходимые ресурсы Azure и развернуть свой код:\n\n```shell\nazd up\n```\n\n> **Примечание.** [API Management]() можно использовать для повышения безопасности и политик на вашем сервере MCP, а[встроенную аутентификацию App Service](https://learn.microsoft.com/en-us/azure/app-service/overview-authentication-authorization) можно использовать для настройки вашего любимого поставщика OAuth, включая Entra.\n\n## Подключитесь к *удаленному* приложению сервера MCP с клиента\n\nВашему клиенту понадобится ключ для вызова новой размещенной конечной точки SSE, которая будет иметь вид `https://.azurewebsites.net/runtime/webhooks/mcp/sse` . Размещенная функция требует системный ключ по умолчанию, который можно получить на [портале](https://learn.microsoft.com/en-us/azure/azure-functions/function-keys-how-to?tabs=azure-portal) или в CLI ( `az functionapp keys list --resource-group --name ` ). Получите системный ключ с именем `mcp_extension` .\n\n### Подключитесь к удаленному серверу MCP в MCP Inspector\n\nДля MCP Inspector вы можете включить ключ в URL:\n\n```plaintext\nhttps://.azurewebsites.net/runtime/webhooks/mcp/sse?code=\n```\n\n### Подключение к удаленному серверу MCP в VS Code - GitHub Copilot\n\nДля GitHub Copilot в VS Code вам следует задать ключ как заголовок `x-functions-key` в `mcp.json` , и вы должны использовать `https://.azurewebsites.net/runtime/webhooks/mcp/sse` для URL. Следующий пример взят из файла `mcp.json` , включенного в этот репозиторий, и использует входные данные для запроса на ввод ключа при запуске сервера из VS Code. Ваш файл `mcp.json` выглядит следующим образом:\n\n```json\n{\n \"inputs\": [\n {\n \"type\": \"promptString\",\n \"id\": \"functions-mcp-extension-system-key\",\n \"description\": \"Azure Functions MCP Extension System Key\",\n \"password\": true\n },\n {\n \"type\": \"promptString\",\n \"id\": \"functionapp-name\",\n \"description\": \"Azure Functions App Name\"\n }\n ],\n \"servers\": {\n \"remote-mcp-function\": {\n \"type\": \"sse\",\n \"url\": \"https://${input:functionapp-name}.azurewebsites.net/runtime/webhooks/mcp/sse\",\n \"headers\": {\n \"x-functions-key\": \"${input:functions-mcp-extension-system-key}\"\n }\n },\n \"local-mcp-function\": {\n \"type\": \"sse\",\n \"url\": \"http://0.0.0.0:7071/runtime/webhooks/mcp/sse\"\n }\n }\n}\n```\n\n1. Нажмите кнопку Пуск на сервере `remote-mcp-function` внутри файла `mcp.json` :\n\n2. Введите имя приложения-функции, созданного на портале Azure, при появлении соответствующего запроса от VS Code.\n\n3. Введите `Azure Functions MCP Extension System Key` в приглашение. Вы можете скопировать его из портала Azure для своего функционального приложения, перейдя в пункт меню Functions, затем App Keys и скопировав ключ `mcp_extension` из System Keys.\n\n4. В режиме чат-агента Copilot введите запрос для запуска инструмента, например, выберите какой-либо код и введите этот запрос.\n\n ```plaintext\n Say Hello\n ```\n\n ```plaintext\n Save this snippet as snippet1 \n ```\n\n ```plaintext\n Retrieve snippet1 and apply to newFile.ts\n ```\n\n## Повторно разверните свой код\n\nВы можете запускать команду `azd up` столько раз, сколько необходимо для подготовки ресурсов Azure и развертывания обновлений кода в вашем функциональном приложении.\n\n> \\[!ПРИМЕЧАНИЕ] Развернутые файлы кода всегда перезаписываются последним пакетом развертывания.\n\n## Очистите ресурсы\n\nЗакончив работу с приложением-функцией и связанными с ним ресурсами, вы можете использовать эту команду, чтобы удалить приложение-функцию и связанные с ним ресурсы из Azure и избежать дальнейших затрат:\n\n```shell\nazd down\n```\n\n## Исходный код\n\nКод функции для конечных точек `getSnippet` и `saveSnippet` определен в файлах TypeScript в каталоге `src` . Аннотации функций MCP раскрывают эти функции как инструменты сервера MCP.\n\nНиже показан код для нескольких примеров сервера MCP (получение строки, получение объекта, сохранение объекта):\n\n```typescript\n// Hello function - responds with hello message\nexport async function mcpToolHello(context: InvocationContext): Promise {\n return \"Hello I am MCP Tool!\";\n}\n\n// Register the hello tool\napp.mcpTool('hello', {\n toolName: 'hello',\n description: 'Simple hello world MCP Tool that responses with a hello message.',\n handler: mcpToolHello\n});\n\n// GetSnippet function - retrieves a snippet by name\nexport async function getSnippet(_message: unknown, context: InvocationContext): Promise {\n console.info('Getting snippet');\n \n // Get snippet name from the tool arguments\n const mcptoolargs = context.triggerMetadata.mcptoolargs as { snippetname?: string };\n const snippetName = mcptoolargs?.snippetname;\n\n console.info(`Snippet name: ${snippetName}`);\n \n if (!snippetName) {\n return \"No snippet name provided\";\n }\n \n // Get the content from blob binding - properly retrieving from extraInputs\n const snippetContent = context.extraInputs.get(blobInputBinding);\n \n if (!snippetContent) {\n return `Snippet '${snippetName}' not found`;\n }\n \n console.info(`Retrieved snippet: ${snippetName}`);\n return snippetContent as string;\n}\n\n\n// Register the GetSnippet tool\napp.mcpTool('getsnippet', {\n toolName: GET_SNIPPET_TOOL_NAME,\n description: GET_SNIPPET_TOOL_DESCRIPTION,\n toolProperties: [\n {\n propertyName: SNIPPET_NAME_PROPERTY_NAME,\n propertyValue: PROPERTY_TYPE,\n description: SNIPPET_NAME_PROPERTY_DESCRIPTION,\n }\n ],\n extraInputs: [blobInputBinding],\n handler: getSnippet\n});\n\n// SaveSnippet function - saves a snippet with a name\nexport async function saveSnippet(_message: unknown, context: InvocationContext): Promise {\n console.info('Saving snippet');\n \n // Get snippet name and content from the tool arguments\n const mcptoolargs = context.triggerMetadata.mcptoolargs as { \n snippetname?: string;\n snippet?: string;\n };\n \n const snippetName = mcptoolargs?.snippetname;\n const snippet = mcptoolargs?.snippet;\n \n if (!snippetName) {\n return \"No snippet name provided\";\n }\n \n if (!snippet) {\n return \"No snippet content provided\";\n }\n \n // Save the snippet to blob storage using the output binding\n context.extraOutputs.set(blobOutputBinding, snippet);\n \n console.info(`Saved snippet: ${snippetName}`);\n return snippet;\n}\n\n// Register the SaveSnippet tool\napp.mcpTool('savesnippet', {\n toolName: SAVE_SNIPPET_TOOL_NAME,\n description: SAVE_SNIPPET_TOOL_DESCRIPTION,\n toolProperties: [\n {\n propertyName: SNIPPET_NAME_PROPERTY_NAME,\n propertyValue: PROPERTY_TYPE,\n description: SNIPPET_NAME_PROPERTY_DESCRIPTION,\n },\n {\n propertyName: SNIPPET_PROPERTY_NAME,\n propertyValue: PROPERTY_TYPE,\n description: SNIPPET_PROPERTY_DESCRIPTION,\n }\n ],\n extraOutputs: [blobOutputBinding],\n handler: saveSnippet\n});\n```\n\nОбратите внимание, что файл `host.json` также содержит ссылку на экспериментальный пакет, который необходим для приложений, использующих эту функцию:\n\n```json\n\"extensionBundle\": {\n \"id\": \"Microsoft.Azure.Functions.ExtensionBundle.Experimental\",\n \"version\": \"[4.*, 5.0.0)\"\n}\n```\n\n## Следующие шаги\n\n* Добавьте [API-управление](https://aka.ms/mcp-remote-apim-auth) на свой сервер MCP (аутентификация, шлюз, политики и многое другое!)\n* Добавьте[встроенную аутентификацию](https://learn.microsoft.com/en-us/azure/app-service/overview-authentication-authorization) на ваш сервер MCP\n* Включите VNET с помощью флага VNET\\_ENABLED=true\n* Узнайте больше о [связанных с MCP инициативах от Microsoft](https://github.com/microsoft/mcp/tree/main/Resources)","Russian","ru-RU",{"_27":74,"_31":78,"_36":79},{"_29":75,"_31":76,"_33":77},"使用 Azure Functions 的基于云的自定义 MCP 服务器，可通过密钥、HTTPS、OAuth 和网络隔离选项通过安全通信保存和检索代码片段。","远程 MCP 服务器","# 使用 Azure Functions (Node.js/TypeScript) 开始使用远程 MCP 服务器\n\n这是一个快速入门模板，可帮助您使用 Azure Functions 轻松构建自定义远程 MCP 服务器并将其部署到云端。您可以在本地计算机上进行克隆/恢复/运行并进行调试，然后几分钟内`azd up`其部署到云端。MCP 服务器采用密钥和 HTTPS 协议进行安全保护，并支持使用 EasyAuth 和/或 API Management 进行 OAuth 授权，以及使用 VNET 进行网络隔离。\n\n**观看视频概述**\n\n如果您正在寻找更多语言的示例，请查看[.NET/C#](https://github.com/Azure-Samples/remote-mcp-functions-dotnet)和[Python](https://github.com/Azure-Samples/remote-mcp-functions-python)版本。\n\n[![在 GitHub Codespaces 中打开](https://github.com/codespaces/badge.svg)](https://codespaces.new/Azure-Samples/remote-mcp-functions-typescript)\n\n以下是使用 Azure Functions 的远程 MCP 服务器的架构图：\n\n![架构图](architecture-diagram.png)\n\n## 先决条件\n\n* [Node.js](https://nodejs.org/en/download/)版本 18 或更高版本\n* [Azure Functions 核心工具](https://learn.microsoft.com/azure/azure-functions/functions-run-local?pivots=programming-language-javascript#install-the-azure-functions-core-tools)>= `4.0.7030`\n* [Azure 开发人员 CLI](https://aka.ms/azd)\n* 要使用 Visual Studio Code 在本地运行和调试：\n * [Visual Studio 代码](https://code.visualstudio.com/)\n * [Azure Functions 扩展](https://marketplace.visualstudio.com/items?itemName=ms-azuretools.vscode-azurefunctions)\n* [Docker](https://www.docker.com/)运行 Azurite，Azure 存储模拟器（可选）\n\n## 准备本地环境\n\n此特定示例需要 Azure 存储模拟器，因为我们将从 Blob 存储中保存和获取片段。\n\n1. 启动 Azurite\n\n ```shell\n docker run -p 10000:10000 -p 10001:10001 -p 10002:10002 \\\n mcr.microsoft.com/azure-storage/azurite\n ```\n\n> \\*\\*请注意，\\*\\*如果您使用来自 VS Code 扩展的 Azurite，则需要运行`Azurite: Start` ，否则您将看到错误。\n\n## 从终端本地运行 MCP 服务器\n\n1. 安装依赖项\n\n ```shell\n npm install\n ```\n\n2. 构建项目\n\n ```shell\n npm run build\n ```\n\n3. 在本地启动 Functions 主机：\n\n ```shell\n func start\n ```\n\n> \\*\\*注意：\\*\\*默认情况下，这将使用 webhooks 路由： `/runtime/webhooks/mcp/sse` 。稍后我们将在 Azure 中使用它来设置客户端/主机调用的密钥： `/runtime/webhooks/mcp/sse?code=`\n\n## 在客户端/主机中使用*本地*MCP 服务器\n\n### VS Code - Copilot 编辑\n\n1. 从命令面板**添加 MCP 服务器**并将 URL 添加到正在运行的函数应用程序的 SSE 端点：\n\n ```shell\n http://0.0.0.0:7071/runtime/webhooks/mcp/sse\n ```\n\n2. 选择\\*\\*HTTP（服务器发送事件）\\*\\*作为要添加的 MCP 服务器类型。\n\n3. 输入正在运行的函数应用的 SSE 终结点的 URL\n\n4. 输入服务器 ID。（可以是任何您想要的名称）\n\n5. 选择是否要在用户设置（适用于所有应用程序）或工作区设置（仅适用于此应用程序）中运行此功能\n\n6. 从命令面板**列出 MCP 服务器**并启动服务器。上一步可能已经启动了您��本地服务器。如果是这样，您可以跳过此步骤。\n\n7. 在 Copilot 聊天代理模式下输入提示来触发该工具，例如，选择一些代码并输入此提示\n\n ```plaintext\n Say Hello\n ```\n\n ```plaintext\n Save this snippet as snippet1 \n ```\n\n ```plaintext\n Retrieve snippet1 and apply to newFile.ts\n ```\n\n8. 当系统提示运行该工具时，单击\\*\\*“继续”\\*\\*表示同意。\n\n9. 完成后，在终端窗口中按 Ctrl+C 停止`func.exe`主机进程，然后从命令面板**列出 MCP 服务器**并停止本地服务器。\n\n### MCP 检查器\n\n1. 在**新的终端窗口**中，安装并运行 MCP Inspector\n\n ```shell\n npx @modelcontextprotocol/inspector node build/index.js\n ```\n\n2. 如果先前已停止函数应用，请在本地启动 Functions 主机：\n\n ```shell\n func start\n ```\n\n3. 按住 CTRL 键并单击，即可从应用程序显示的 URL 加载 MCP Inspector Web 应用程序（例如）\n\n4. 将传输类型设置为`SSE`\n\n5. 将 URL 设置为正在运行的 Function App 的 SSE 端点并**进行连接**：\n\n ```shell\n http://0.0.0.0:7071/runtime/webhooks/mcp/sse\n ```\n\n6. **列出工具**。单击某个工具并**运行工具**。\n\n7. 完成后，在终端窗口中按 Ctrl+C 停止`func.exe`主机进程，在终端窗口中按 Ctrl+C 停止`@modelcontextprotocol/inspector`主机进程。\n\n## 部署到 Azure 以进行远程 MCP\n\n或者，您可以选择加入示例中使用的 VNet。（如果选择此选项，请在`azd up`之前执行此操作）\n\n```bash\nazd env set VNET_ENABLED true\n```\n\n运行此[azd](https://aka.ms/azd)命令来配置函数应用程序及其所需的 Azure 资源，并部署代码：\n\n```shell\nazd up\n```\n\n> **注意**[API 管理]()可用于改进 MCP 服务器的安全性和策略，而[应用服务内置身份验证](https://learn.microsoft.com/en-us/azure/app-service/overview-authentication-authorization)可用于设置您最喜欢的 OAuth 提供程序（包括 Entra）。\n\n## 从客户端连接到*远程*MCP 服务器函数应用程序\n\n客户端需要一个密钥才能调用新的托管 SSE 终结点，其格式为`https://.azurewebsites.net/runtime/webhooks/mcp/sse` 。默认情况下，托管函数需要一个系统密钥，该密钥可从[门户](https://learn.microsoft.com/en-us/azure/azure-functions/function-keys-how-to?tabs=azure-portal)或 CLI ( `az functionapp keys list --resource-group --name ` ) 获取。获取名为`mcp_extension`的系统密钥。\n\n### 在 MCP Inspector 中连接到远程 MCP 服务器\n\n对于 MCP Inspector，您可以在 URL 中包含密钥：\n\n```plaintext\nhttps://.azurewebsites.net/runtime/webhooks/mcp/sse?code=\n```\n\n### 在 VS Code 中连接到远程 MCP 服务器 - GitHub Copilot\n\n对于 VS Code 中的 GitHub Copilot，应将密钥设置为`mcp.json`中的`x-functions-key`标头，并使用`https://.azurewebsites.net/runtime/webhooks/mcp/sse`作为 URL。以下示例取自此存储库中包含的`mcp.json`文件，并在您从 VS Code 启动服务器时使用输入提示您提供密钥。您的`mcp.json`文件如下所示：\n\n```json\n{\n \"inputs\": [\n {\n \"type\": \"promptString\",\n \"id\": \"functions-mcp-extension-system-key\",\n \"description\": \"Azure Functions MCP Extension System Key\",\n \"password\": true\n },\n {\n \"type\": \"promptString\",\n \"id\": \"functionapp-name\",\n \"description\": \"Azure Functions App Name\"\n }\n ],\n \"servers\": {\n \"remote-mcp-function\": {\n \"type\": \"sse\",\n \"url\": \"https://${input:functionapp-name}.azurewebsites.net/runtime/webhooks/mcp/sse\",\n \"headers\": {\n \"x-functions-key\": \"${input:functions-mcp-extension-system-key}\"\n }\n },\n \"local-mcp-function\": {\n \"type\": \"sse\",\n \"url\": \"http://0.0.0.0:7071/runtime/webhooks/mcp/sse\"\n }\n }\n}\n```\n\n1. 在服务器`remote-mcp-function`中，单击`mcp.json`文件内的 Start ：\n\n2. 当 VS Code 提示时，输入您在 Azure 门户中创建的函数应用程序的名称。\n\n3. 在提示符中输入`Azure Functions MCP Extension System Key` 。您可以从 Azure 门户为您的函数应用复制此密钥，方法是转到“函数”菜单项，然后选择“应用密钥”，然后从“系统密钥”中复制`mcp_extension`密钥。\n\n4. 在 Copilot 聊天代理模式下输入提示来触发该工具，例如，选择一些代码并输入此提示\n\n ```plaintext\n Say Hello\n ```\n\n ```plaintext\n Save this snippet as snippet1 \n ```\n\n ```plaintext\n Retrieve snippet1 and apply to newFile.ts\n ```\n\n## 重新部署代码\n\n您可以根据需要多次运行`azd up`命令，以配置 Azure 资源并将代码更新部署到函数应用程序。\n\n> \\[!NOTE] 已部署的代码文件始终会被最新的部署包覆盖。\n\n## 清理资源\n\n使用完函数应用和相关资源后，可以使用此命令从 Azure 中删除函数应用及其相关资源，并避免产生任何进一步的费用：\n\n```shell\nazd down\n```\n\n## 源代码\n\n`getSnippet`和`saveSnippet`端点的函数代码定义在`src`目录下的 TypeScript 文件中。MCP 函数注释将这些函数公开为 MCP 服务器工具。\n\n这显示了一些 MCP 服务器示例的代码（获取字符串、获取对象、保存对象）：\n\n```typescript\n// Hello function - responds with hello message\nexport async function mcpToolHello(context: InvocationContext): Promise {\n return \"Hello I am MCP Tool!\";\n}\n\n// Register the hello tool\napp.mcpTool('hello', {\n toolName: 'hello',\n description: 'Simple hello world MCP Tool that responses with a hello message.',\n handler: mcpToolHello\n});\n\n// GetSnippet function - retrieves a snippet by name\nexport async function getSnippet(_message: unknown, context: InvocationContext): Promise {\n console.info('Getting snippet');\n \n // Get snippet name from the tool arguments\n const mcptoolargs = context.triggerMetadata.mcptoolargs as { snippetname?: string };\n const snippetName = mcptoolargs?.snippetname;\n\n console.info(`Snippet name: ${snippetName}`);\n \n if (!snippetName) {\n return \"No snippet name provided\";\n }\n \n // Get the content from blob binding - properly retrieving from extraInputs\n const snippetContent = context.extraInputs.get(blobInputBinding);\n \n if (!snippetContent) {\n return `Snippet '${snippetName}' not found`;\n }\n \n console.info(`Retrieved snippet: ${snippetName}`);\n return snippetContent as string;\n}\n\n\n// Register the GetSnippet tool\napp.mcpTool('getsnippet', {\n toolName: GET_SNIPPET_TOOL_NAME,\n description: GET_SNIPPET_TOOL_DESCRIPTION,\n toolProperties: [\n {\n propertyName: SNIPPET_NAME_PROPERTY_NAME,\n propertyValue: PROPERTY_TYPE,\n description: SNIPPET_NAME_PROPERTY_DESCRIPTION,\n }\n ],\n extraInputs: [blobInputBinding],\n handler: getSnippet\n});\n\n// SaveSnippet function - saves a snippet with a name\nexport async function saveSnippet(_message: unknown, context: InvocationContext): Promise {\n console.info('Saving snippet');\n \n // Get snippet name and content from the tool arguments\n const mcptoolargs = context.triggerMetadata.mcptoolargs as { \n snippetname?: string;\n snippet?: string;\n };\n \n const snippetName = mcptoolargs?.snippetname;\n const snippet = mcptoolargs?.snippet;\n \n if (!snippetName) {\n return \"No snippet name provided\";\n }\n \n if (!snippet) {\n return \"No snippet content provided\";\n }\n \n // Save the snippet to blob storage using the output binding\n context.extraOutputs.set(blobOutputBinding, snippet);\n \n console.info(`Saved snippet: ${snippetName}`);\n return snippet;\n}\n\n// Register the SaveSnippet tool\napp.mcpTool('savesnippet', {\n toolName: SAVE_SNIPPET_TOOL_NAME,\n description: SAVE_SNIPPET_TOOL_DESCRIPTION,\n toolProperties: [\n {\n propertyName: SNIPPET_NAME_PROPERTY_NAME,\n propertyValue: PROPERTY_TYPE,\n description: SNIPPET_NAME_PROPERTY_DESCRIPTION,\n },\n {\n propertyName: SNIPPET_PROPERTY_NAME,\n propertyValue: PROPERTY_TYPE,\n description: SNIPPET_PROPERTY_DESCRIPTION,\n }\n ],\n extraOutputs: [blobOutputBinding],\n handler: saveSnippet\n});\n```\n\n请注意， `host.json`文件还包含对实验包的引用，这是使用此功能的应用程序所必需的：\n\n```json\n\"extensionBundle\": {\n \"id\": \"Microsoft.Azure.Functions.ExtensionBundle.Experimental\",\n \"version\": \"[4.*, 5.0.0)\"\n}\n```\n\n## 后续步骤\n\n* 将[API 管理](https://aka.ms/mcp-remote-apim-auth)添加到您的 MCP 服务器（授权、网关、策略等！）\n* 为您的 MCP 服务器添加[内置身份验证](https://learn.microsoft.com/en-us/azure/app-service/overview-authentication-authorization)\n* 使用 VNET\\_ENABLED=true 标志启用 VNET\n* 详细了解[Microsoft 的 MCP 相关工作](https://github.com/microsoft/mcp/tree/main/Resources)","Chinese","zh-CN","githubUser","locale",{"_83":84,"_13":85,"_92":40,"_93":40,"_16":94,"_95":96,"_31":41,"_143":144,"_151":-5,"_152":-5,"_153":154,"_198":199,"_204":-5,"_103":179,"_205":206,"_106":207,"_208":209},"addedAt","2025-04-03T23:01:12.286740Z",[86,87,88,89,90,91],"category:cloud-platforms","category:code-execution","environment:network","hosting:hybrid","language:bicep","license_grade:a","descriptionMarkdown","descriptionPlainText",4933,"integrations",[97,108,115,122,129,136],{"_98":99,"_29":105,"_106":107},"brand",{"_100":101,"_31":102,"_103":104},"logoSvgPath","M13.983 11.078h2.119a.186.186 0 00.186-.185V9.006a.186.186 0 00-.186-.186h-2.119a.185.185 0 00-.185.185v1.888c0 .102.083.185.185.185m-2.954-5.43h2.118a.186.186 0 00.186-.186V3.574a.186.186 0 00-.186-.185h-2.118a.185.185 0 00-.185.185v1.888c0 .102.082.185.185.185m0 2.716h2.118a.187.187 0 00.186-.186V6.29a.186.186 0 00-.186-.185h-2.118a.185.185 0 00-.185.185v1.887c0 .102.082.185.185.186m-2.93 0h2.12a.186.186 0 00.184-.186V6.29a.185.185 0 00-.185-.185H8.1a.185.185 0 00-.185.185v1.887c0 .102.083.185.185.186m-2.964 0h2.119a.186.186 0 00.185-.186V6.29a.185.185 0 00-.185-.185H5.136a.186.186 0 00-.186.185v1.887c0 .102.084.185.186.186m5.893 2.715h2.118a.186.186 0 00.186-.185V9.006a.186.186 0 00-.186-.186h-2.118a.185.185 0 00-.185.185v1.888c0 .102.082.185.185.185m-2.93 0h2.12a.185.185 0 00.184-.185V9.006a.185.185 0 00-.184-.186h-2.12a.185.185 0 00-.184.185v1.888c0 .102.083.185.185.185m-2.964 0h2.119a.185.185 0 00.185-.185V9.006a.185.185 0 00-.184-.186h-2.12a.186.186 0 00-.186.186v1.887c0 .102.084.185.186.185m-2.92 0h2.12a.185.185 0 00.184-.185V9.006a.185.185 0 00-.184-.186h-2.12a.185.185 0 00-.184.185v1.888c0 .102.082.185.185.185M23.763 9.89c-.065-.051-.672-.51-1.954-.51-.338.001-.676.03-1.01.087-.248-1.7-1.653-2.53-1.716-2.566l-.344-.199-.226.327c-.284.438-.49.922-.612 1.43-.23.97-.09 1.882.403 2.661-.595.332-1.55.413-1.744.42H.751a.751.751 0 00-.75.748 11.376 11.376 0 00.692 4.062c.545 1.428 1.355 2.48 2.41 3.124 1.18.723 3.1 1.137 5.275 1.137.983.003 1.963-.086 2.93-.266a12.248 12.248 0 003.823-1.389c.98-.567 1.86-1.288 2.61-2.136 1.252-1.418 1.998-2.997 2.553-4.4h.221c1.372 0 2.215-.549 2.68-1.009.309-.293.55-.65.707-1.046l.098-.288Z","Docker","slug","docker","Supports running Azure Storage Emulator through Docker during local development for testing blob storage operations.","uid","hjceqc6hb7",{"_98":109,"_29":113,"_106":114},{"_100":110,"_31":111,"_103":112},"M12 .297c-6.63 0-12 5.373-12 12 0 5.303 3.438 9.8 8.205 11.385.6.113.82-.258.82-.577 0-.285-.01-1.04-.015-2.04-3.338.724-4.042-1.61-4.042-1.61C4.422 18.07 3.633 17.7 3.633 17.7c-1.087-.744.084-.729.084-.729 1.205.084 1.838 1.236 1.838 1.236 1.07 1.835 2.809 1.305 3.495.998.108-.776.417-1.305.76-1.605-2.665-.3-5.466-1.332-5.466-5.93 0-1.31.465-2.38 1.235-3.22-.135-.303-.54-1.523.105-3.176 0 0 1.005-.322 3.3 1.23.96-.267 1.98-.399 3-.405 1.02.006 2.04.138 3 .405 2.28-1.552 3.285-1.23 3.285-1.23.645 1.653.24 2.873.12 3.176.765.84 1.23 1.91 1.23 3.22 0 4.61-2.805 5.625-5.475 5.92.42.36.81 1.096.81 2.22 0 1.606-.015 2.896-.015 3.286 0 .315.21.69.825.57C20.565 22.092 24 17.592 24 12.297c0-6.627-5.373-12-12-12","GitHub","github","Supports development in GitHub Codespaces for a ready-to-use development environment without local setup.","xmp6fzc99b",{"_98":116,"_29":120,"_106":121},{"_100":117,"_31":118,"_103":119},"M24 8.77h-2.468v7.565h-1.425V8.77h-2.462V7.53H24zm-6.852 7.565h-4.821V7.53h4.63v1.24h-3.205v2.494h2.953v1.234h-2.953v2.604h3.396zm-6.708 0H8.882L4.78 9.863a2.896 2.896 0 0 1-.258-.51h-.036c.032.189.048.592.048 1.21v5.772H3.157V7.53h1.659l3.965 6.32c.167.261.275.442.323.54h.024c-.04-.233-.06-.629-.06-1.185V7.529h1.372zm-8.703-.693a.868.829 0 0 1-.869.829.868.829 0 0 1-.868-.83.868.829 0 0 1 .868-.828.868.829 0 0 1 .869.829Z",".NET","dotnet","Offers a similar MCP server implementation available in .NET/C# for developers preferring that ecosystem.","s4sv78qqea",{"_98":123,"_29":127,"_106":128},{"_100":124,"_31":125,"_103":126},"M11.998,24c-0.321,0-0.641-0.084-0.922-0.247l-2.936-1.737c-0.438-0.245-0.224-0.332-0.08-0.383 c0.585-0.203,0.703-0.25,1.328-0.604c0.065-0.037,0.151-0.023,0.218,0.017l2.256,1.339c0.082,0.045,0.197,0.045,0.272,0l8.795-5.076 c0.082-0.047,0.134-0.141,0.134-0.238V6.921c0-0.099-0.053-0.192-0.137-0.242l-8.791-5.072c-0.081-0.047-0.189-0.047-0.271,0 L3.075,6.68C2.99,6.729,2.936,6.825,2.936,6.921v10.15c0,0.097,0.054,0.189,0.139,0.235l2.409,1.392 c1.307,0.654,2.108-0.116,2.108-0.89V7.787c0-0.142,0.114-0.253,0.256-0.253h1.115c0.139,0,0.255,0.112,0.255,0.253v10.021 c0,1.745-0.95,2.745-2.604,2.745c-0.508,0-0.909,0-2.026-0.551L2.28,18.675c-0.57-0.329-0.922-0.945-0.922-1.604V6.921 c0-0.659,0.353-1.275,0.922-1.603l8.795-5.082c0.557-0.315,1.296-0.315,1.848,0l8.794,5.082c0.57,0.329,0.924,0.944,0.924,1.603 v10.15c0,0.659-0.354,1.273-0.924,1.604l-8.794,5.078C12.643,23.916,12.324,24,11.998,24z M19.099,13.993 c0-1.9-1.284-2.406-3.987-2.763c-2.731-0.361-3.009-0.548-3.009-1.187c0-0.528,0.235-1.233,2.258-1.233 c1.807,0,2.473,0.389,2.747,1.607c0.024,0.115,0.129,0.199,0.247,0.199h1.141c0.071,0,0.138-0.031,0.186-0.081 c0.048-0.054,0.074-0.123,0.067-0.196c-0.177-2.098-1.571-3.076-4.388-3.076c-2.508,0-4.004,1.058-4.004,2.833 c0,1.925,1.488,2.457,3.895,2.695c2.88,0.282,3.103,0.703,3.103,1.269c0,0.983-0.789,1.402-2.642,1.402 c-2.327,0-2.839-0.584-3.011-1.742c-0.02-0.124-0.126-0.215-0.253-0.215h-1.137c-0.141,0-0.254,0.112-0.254,0.253 c0,1.482,0.806,3.248,4.655,3.248C17.501,17.007,19.099,15.91,19.099,13.993z","Node.js","node-dot-js","Primary implementation language for this MCP server, allowing Node.js developers to build and deploy custom remote MCP tools.","69oajkmmir",{"_98":130,"_29":134,"_106":135},{"_100":131,"_31":132,"_103":133},"M14.25.18l.9.2.73.26.59.3.45.32.34.34.25.34.16.33.1.3.04.26.02.2-.01.13V8.5l-.05.63-.13.55-.21.46-.26.38-.3.31-.33.25-.35.19-.35.14-.33.1-.3.07-.26.04-.21.02H8.77l-.69.05-.59.14-.5.22-.41.27-.33.32-.27.35-.2.36-.15.37-.1.35-.07.32-.04.27-.02.21v3.06H3.17l-.21-.03-.28-.07-.32-.12-.35-.18-.36-.26-.36-.36-.35-.46-.32-.59-.28-.73-.21-.88-.14-1.05-.05-1.23.06-1.22.16-1.04.24-.87.32-.71.36-.57.4-.44.42-.33.42-.24.4-.16.36-.1.32-.05.24-.01h.16l.06.01h8.16v-.83H6.18l-.01-2.75-.02-.37.05-.34.11-.31.17-.28.25-.26.31-.23.38-.2.44-.18.51-.15.58-.12.64-.1.71-.06.77-.04.84-.02 1.27.05zm-6.3 1.98l-.23.33-.08.41.08.41.23.34.33.22.41.09.41-.09.33-.22.23-.34.08-.41-.08-.41-.23-.33-.33-.22-.41-.09-.41.09zm13.09 3.95l.28.06.32.12.35.18.36.27.36.35.35.47.32.59.28.73.21.88.14 1.04.05 1.23-.06 1.23-.16 1.04-.24.86-.32.71-.36.57-.4.45-.42.33-.42.24-.4.16-.36.09-.32.05-.24.02-.16-.01h-8.22v.82h5.84l.01 2.76.02.36-.05.34-.11.31-.17.29-.25.25-.31.24-.38.2-.44.17-.51.15-.58.13-.64.09-.71.07-.77.04-.84.01-1.27-.04-1.07-.14-.9-.2-.73-.25-.59-.3-.45-.33-.34-.34-.25-.34-.16-.33-.1-.3-.04-.25-.02-.2.01-.13v-5.34l.05-.64.13-.54.21-.46.26-.38.3-.32.33-.24.35-.2.35-.14.33-.1.3-.06.26-.04.21-.02.13-.01h5.84l.69-.05.59-.14.5-.21.41-.28.33-.32.27-.35.2-.36.15-.36.1-.35.07-.32.04-.28.02-.21V6.07h2.09l.14.01zm-6.47 14.25l-.23.33-.08.41.08.41.23.33.33.23.41.08.41-.08.33-.23.23-.33.08-.41-.08-.41-.23-.33-.33-.23-.41-.08-.41.08z","Python","python","Provides an alternative MCP server implementation in Python for developers working in that language.","gkx1tndeh2",{"_98":137,"_29":141,"_106":142},{"_100":138,"_31":139,"_103":140},"M1.125 0C.502 0 0 .502 0 1.125v21.75C0 23.498.502 24 1.125 24h21.75c.623 0 1.125-.502 1.125-1.125V1.125C24 .502 23.498 0 22.875 0zm17.363 9.75c.612 0 1.154.037 1.627.111a6.38 6.38 0 0 1 1.306.34v2.458a3.95 3.95 0 0 0-.643-.361 5.093 5.093 0 0 0-.717-.26 5.453 5.453 0 0 0-1.426-.2c-.3 0-.573.028-.819.086a2.1 2.1 0 0 0-.623.242c-.17.104-.3.229-.393.374a.888.888 0 0 0-.14.49c0 .196.053.373.156.529.104.156.252.304.443.444s.423.276.696.41c.273.135.582.274.926.416.47.197.892.407 1.266.628.374.222.695.473.963.753.268.279.472.598.614.957.142.359.214.776.214 1.253 0 .657-.125 1.21-.373 1.656a3.033 3.033 0 0 1-1.012 1.085 4.38 4.38 0 0 1-1.487.596c-.566.12-1.163.18-1.79.18a9.916 9.916 0 0 1-1.84-.164 5.544 5.544 0 0 1-1.512-.493v-2.63a5.033 5.033 0 0 0 3.237 1.2c.333 0 .624-.03.872-.09.249-.06.456-.144.623-.25.166-.108.29-.234.373-.38a1.023 1.023 0 0 0-.074-1.089 2.12 2.12 0 0 0-.537-.5 5.597 5.597 0 0 0-.807-.444 27.72 27.72 0 0 0-1.007-.436c-.918-.383-1.602-.852-2.053-1.405-.45-.553-.676-1.222-.676-2.005 0-.614.123-1.141.369-1.582.246-.441.58-.804 1.004-1.089a4.494 4.494 0 0 1 1.47-.629 7.536 7.536 0 0 1 1.77-.201zm-15.113.188h9.563v2.166H9.506v9.646H6.789v-9.646H3.375z","TypeScript","typescript","Enables development of type-safe MCP server tools using TypeScript for improved code quality and developer experience.","2agz4gb3w1","namespace",{"_145":146,"_103":150},"owner",{"_147":148},"githubAccount",{"_149":150},"login","Azure-Samples","readmeMarkdown","redditPostExternalId","repository",{"_155":156,"_163":164,"_168":169,"_189":-5,"_190":191,"_192":-5,"_193":194},"argumentsJsonSchema",{"_157":158,"_159":160,"_161":162},"properties",{},"required",[],"type","object","githubProject",{"_33":165,"_166":167},"\n# Getting Started with Remote MCP Servers using Azure Functions (Node.js/TypeScript)\n\nThis is a quickstart template to easily build and deploy a custom remote MCP server to the cloud using Azure functions. You can clone/restore/run on your local machine with debugging, and `azd up` to have it in the cloud in a couple minutes. The MCP server is secured by design using keys and HTTPs, and allows more options for OAuth using EasyAuth and/or API Management as well as network isolation using VNET. \n\n**Watch the video overview**\n\n\n $\"Watch$ \n\n\nIf you're looking for this sample in more languages check out the [.NET/C#](https://github.com/Azure-Samples/remote-mcp-functions-dotnet) and [Python](https://github.com/Azure-Samples/remote-mcp-functions-python) versions.\n\n[![Open in GitHub Codespaces](https://github.com/codespaces/badge.svg)](https://codespaces.new/Azure-Samples/remote-mcp-functions-typescript)\n\nBelow is the architecture diagram for the Remote MCP Server using Azure Functions:\n\n![Architecture Diagram](architecture-diagram.png)\n\n## Prerequisites\n\n+ [Node.js](https://nodejs.org/en/download/) version 18 or higher\n+ [Azure Functions Core Tools](https://learn.microsoft.com/azure/azure-functions/functions-run-local?pivots=programming-language-javascript#install-the-azure-functions-core-tools) >= `4.0.7030`\n+ [Azure Developer CLI](https://aka.ms/azd)\n+ To use Visual Studio Code to run and debug locally:\n + [Visual Studio Code](https://code.visualstudio.com/)\n + [Azure Functions extension](https://marketplace.visualstudio.com/items?itemName=ms-azuretools.vscode-azurefunctions)\n+ [Docker](https://www.docker.com/) to run Azurite, the Azure Storage Emulator (optional)\n\n## Prepare your local environment\n\nAn Azure Storage Emulator is needed for this particular sample because we will save and get snippets from blob storage. \n\n1. Start Azurite\n\n ```shell\n docker run -p 10000:10000 -p 10001:10001 -p 10002:10002 \\\n mcr.microsoft.com/azure-storage/azurite\n ```\n\n>**Note** if you use Azurite coming from VS Code extension you need to run `Azurite: Start` now or you will see errors.\n\n## Run your MCP Server locally from the terminal\n\n1. Install dependencies\n ```shell\n npm install\n ```\n\n1. Build the project\n ```shell\n npm run build\n ```\n\n1. Start the Functions host locally:\n ```shell\n func start\n ```\n\n> **Note** by default this will use the webhooks route: `/runtime/webhooks/mcp/sse`. Later we will use this in Azure to set the key on client/host calls: `/runtime/webhooks/mcp/sse?code=`\n\n## Use the *local* MCP server from within a client/host\n\n### VS Code - Copilot Edits\n\n1. **Add MCP Server** from command palette and add URL to your running Function app's SSE endpoint:\n ```shell\n http://0.0.0.0:7071/runtime/webhooks/mcp/sse\n ```\n\n1. Select **HTTP (Server-Sent-Events)** for the type of MCP server to add.\n1. Enter the URL to your running function app's SSE endpoint\n1. Enter the server ID. (This can be any name you want)\n1. Choose if you want to run this in your User settings (available to all apps for you) or to your Workspace settings (available to this app, only)\n1. **List MCP Servers** from command palette and start the server. The previous step may have already started your local server. If so, you can skip this step.\n1. In Copilot chat agent mode enter a prompt to trigger the tool, e.g., select some code and enter this prompt\n\n ```plaintext\n Say Hello\n ```\n\n ```plaintext\n Save this snippet as snippet1 \n ```\n\n ```plaintext\n Retrieve snippet1 and apply to newFile.ts\n ```\n1. When prompted to run the tool, consent by clicking **Continue**\n\n1. When you're done, press Ctrl+C in the terminal window to stop the `func.exe` host process, and **List MCP Servers** from command palette and stop the local server.\n\n### MCP Inspector\n\n1. In a **new terminal window**, install and run MCP Inspector\n\n ```shell\n npx @modelcontextprotocol/inspector node build/index.js\n ```\n \n1. If you stopped your function app previously, start the Functions host locally:\n\n ```shell\n func start\n ```\n\n1. CTRL click to load the MCP Inspector web app from the URL displayed by the app (e.g. http://0.0.0.0:5173/#resources)\n1. Set the transport type to `SSE` \n1. Set the URL to your running Function app's SSE endpoint and **Connect**:\n ```shell\n http://0.0.0.0:7071/runtime/webhooks/mcp/sse\n ```\n1. **List Tools**. Click on a tool and **Run Tool**. \n\n1. When you're done, press Ctrl+C in the terminal window to stop the `func.exe` host process, and press Ctrl+C in the terminal window to stop the `@modelcontextprotocol/inspector` host process.\n\n## Verify local blob storage in Azurite\n\nAfter testing the snippet save functionality locally, you can verify that blobs are being stored correctly in your local Azurite storage emulator.\n\n### Using Azure Storage Explorer\n\n1. Open Azure Storage Explorer\n1. In the left panel, expand **Emulator & Attached** → **Storage Accounts** → **(Emulator - Default Ports) (Key)**\n1. Navigate to **Blob Containers** → **snippets**\n1. You should see any saved snippets as blob files in this container\n1. Double-click on any blob to view its contents and verify the snippet data was saved correctly\n\n### Using Azure CLI (Alternative)\n\nIf you prefer using the command line, you can also verify blobs using Azure CLI with the storage emulator:\n\n```shell\n# List blobs in the snippets container\naz storage blob list --container-name snippets --connection-string \"DefaultEndpointsProtocol=http;AccountName=devstoreaccount1;AccountKey=Eby8vdM02xNOcqFlqUwJPLlmEtlCDXJ1OUzFT50uSRZ6IFsuFq2UVErCz4I6tq/K1SZFPTOtr/KBHBeksoGMGw==;BlobEndpoint=http://127.0.0.1:10000/devstoreaccount1;\"\n```\n\n```shell\n# Download a specific blob to view its contents\naz storage blob download --container-name snippets --name --file --connection-string \"DefaultEndpointsProtocol=http;AccountName=devstoreaccount1;AccountKey=Eby8vdM02xNOcqFlqUwJPLlmEtlCDXJ1OUzFT50uSRZ6IFsuFq2UVErCz4I6tq/K1SZFPTOtr/KBHBeksoGMGw==;BlobEndpoint=http://127.0.0.1:10000/devstoreaccount1;\"\n```\n\nThis verification step ensures your MCP server is correctly interacting with the local storage emulator and that the blob storage functionality is working as expected before deploying to Azure.\n\n## Deploy to Azure for Remote MCP\n\nOptionally, you can opt-in to a VNet being used in the sample. (If you choose this, do this before `azd up`)\n\n```bash\nazd env set VNET_ENABLED true\n```\n\nRun this [azd](https://aka.ms/azd) command to provision the function app, with any required Azure resources, and deploy your code:\n\n```shell\nazd up\n```\n\n> **Note** [API Management]() can be used for improved security and policies over your MCP Server, and [App Service built-in authentication](https://learn.microsoft.com/en-us/azure/app-service/overview-authentication-authorization) can be used to set up your favorite OAuth provider including Entra. \n\n## Connect to your *remote* MCP server function app from a client\n\nYour client will need a key in order to invoke the new hosted SSE endpoint, which will be of the form `https://.azurewebsites.net/runtime/webhooks/mcp/sse`. The hosted function requires a system key by default which can be obtained from the [portal](https://learn.microsoft.com/en-us/azure/azure-functions/function-keys-how-to?tabs=azure-portal) or the CLI (`az functionapp keys list --resource-group --name `). Obtain the system key named `mcp_extension`.\n\n### Connect to remote MCP server in MCP Inspector\nFor MCP Inspector, you can include the key in the URL: \n```plaintext\nhttps://.azurewebsites.net/runtime/webhooks/mcp/sse?code=\n```\n\n### Connect to remote MCP server in VS Code - GitHub Copilot\nFor GitHub Copilot within VS Code, you should set the key as the `x-functions-key` header in `mcp.json`, and you would use `https://.azurewebsites.net/runtime/webhooks/mcp/sse` for the URL. The following example is from the `mcp.json` file included in this repository and uses an input to prompt you to provide the key when you start the server from VS Code. Your `mcp.json` file looks like this:\n\n```json\n{\n \"inputs\": [\n {\n \"type\": \"promptString\",\n \"id\": \"functions-mcp-extension-system-key\",\n \"description\": \"Azure Functions MCP Extension System Key\",\n \"password\": true\n },\n {\n \"type\": \"promptString\",\n \"id\": \"functionapp-name\",\n \"description\": \"Azure Functions App Name\"\n }\n ],\n \"servers\": {\n \"remote-mcp-function\": {\n \"type\": \"sse\",\n \"url\": \"https://${input:functionapp-name}.azurewebsites.net/runtime/webhooks/mcp/sse\",\n \"headers\": {\n \"x-functions-key\": \"${input:functions-mcp-extension-system-key}\"\n }\n },\n \"local-mcp-function\": {\n \"type\": \"sse\",\n \"url\": \"http://0.0.0.0:7071/runtime/webhooks/mcp/sse\"\n }\n }\n}\n```\n\n1. Click Start on the server `remote-mcp-function`, inside the `mcp.json` file:\n\n1. Enter the name of the function app that you created in the Azure Portal, when prompted by VS Code.\n\n1. Enter the `Azure Functions MCP Extension System Key` into the prompt. You can copy this from the Azure portal for your function app by going to the Functions menu item, then App Keys, and copying the `mcp_extension` key from the System Keys.\n\n1. In Copilot chat agent mode enter a prompt to trigger the tool, e.g., select some code and enter this prompt\n\n ```plaintext\n Say Hello\n ```\n\n ```plaintext\n Save this snippet as snippet1 \n ```\n\n ```plaintext\n Retrieve snippet1 and apply to newFile.ts\n ```\n\n## Redeploy your code\n\nYou can run the `azd up` command as many times as you need to both provision your Azure resources and deploy code updates to your function app.\n\n>[!NOTE]\n>Deployed code files are always overwritten by the latest deployment package.\n\n## Clean up resources\n\nWhen you're done working with your function app and related resources, you can use this command to delete the function app and its related resources from Azure and avoid incurring any further costs:\n\n```shell\nazd down\n```\n\n## Source Code\n\nThe function code for the `getSnippet` and `saveSnippet` endpoints are defined in the TypeScript files in the `src` directory. The MCP function annotations expose these functions as MCP Server tools.\n\nThis shows the code for a few MCP server examples (get string, get object, save object):\n\n```typescript\n// Hello function - responds with hello message\nexport async function mcpToolHello(context: InvocationContext): Promise {\n return \"Hello I am MCP Tool!\";\n}\n\n// Register the hello tool\napp.mcpTool('hello', {\n toolName: 'hello',\n description: 'Simple hello world MCP Tool that responses with a hello message.',\n handler: mcpToolHello\n});\n\n// GetSnippet function - retrieves a snippet by name\nexport async function getSnippet(_message: unknown, context: InvocationContext): Promise {\n console.info('Getting snippet');\n \n // Get snippet name from the tool arguments\n const mcptoolargs = context.triggerMetadata.mcptoolargs as { snippetname?: string };\n const snippetName = mcptoolargs?.snippetname;\n\n console.info(`Snippet name: ${snippetName}`);\n \n if (!snippetName) {\n return \"No snippet name provided\";\n }\n \n // Get the content from blob binding - properly retrieving from extraInputs\n const snippetContent = context.extraInputs.get(blobInputBinding);\n \n if (!snippetContent) {\n return `Snippet '${snippetName}' not found`;\n }\n \n console.info(`Retrieved snippet: ${snippetName}`);\n return snippetContent as string;\n}\n\n\n// Register the GetSnippet tool\napp.mcpTool('getsnippet', {\n toolName: GET_SNIPPET_TOOL_NAME,\n description: GET_SNIPPET_TOOL_DESCRIPTION,\n toolProperties: [\n {\n propertyName: SNIPPET_NAME_PROPERTY_NAME,\n propertyValue: PROPERTY_TYPE,\n description: SNIPPET_NAME_PROPERTY_DESCRIPTION,\n }\n ],\n extraInputs: [blobInputBinding],\n handler: getSnippet\n});\n\n// SaveSnippet function - saves a snippet with a name\nexport async function saveSnippet(_message: unknown, context: InvocationContext): Promise {\n console.info('Saving snippet');\n \n // Get snippet name and content from the tool arguments\n const mcptoolargs = context.triggerMetadata.mcptoolargs as { \n snippetname?: string;\n snippet?: string;\n };\n \n const snippetName = mcptoolargs?.snippetname;\n const snippet = mcptoolargs?.snippet;\n \n if (!snippetName) {\n return \"No snippet name provided\";\n }\n \n if (!snippet) {\n return \"No snippet content provided\";\n }\n \n // Save the snippet to blob storage using the output binding\n context.extraOutputs.set(blobOutputBinding, snippet);\n \n console.info(`Saved snippet: ${snippetName}`);\n return snippet;\n}\n\n// Register the SaveSnippet tool\napp.mcpTool('savesnippet', {\n toolName: SAVE_SNIPPET_TOOL_NAME,\n description: SAVE_SNIPPET_TOOL_DESCRIPTION,\n toolProperties: [\n {\n propertyName: SNIPPET_NAME_PROPERTY_NAME,\n propertyValue: PROPERTY_TYPE,\n description: SNIPPET_NAME_PROPERTY_DESCRIPTION,\n },\n {\n propertyName: SNIPPET_PROPERTY_NAME,\n propertyValue: PROPERTY_TYPE,\n description: SNIPPET_PROPERTY_DESCRIPTION,\n }\n ],\n extraOutputs: [blobOutputBinding],\n handler: saveSnippet\n});\n```\n\nNote that the `host.json` file also includes a reference to the experimental bundle, which is required for apps using this feature:\n\n```json\n\"extensionBundle\": {\n \"id\": \"Microsoft.Azure.Functions.ExtensionBundle.Experimental\",\n \"version\": \"[4.*, 5.0.0)\"\n}\n```\n\n## Next Steps\n\n- Add [API Management](https://aka.ms/mcp-remote-apim-auth) to your MCP server (auth, gateway, policies, more!)\n- Add [built-in auth](https://learn.microsoft.com/en-us/azure/app-service/overview-authentication-authorization) to your MCP server\n- Enable VNET using VNET_ENABLED=true flag\n- Learn more about [related MCP efforts from Microsoft](https://github.com/microsoft/mcp/tree/main/Resources)\n\n","url","https://github.com/Azure-Samples/remote-mcp-functions-typescript","githubRepository",{"_170":171,"_172":173,"_174":175,"_16":176,"_177":178,"_31":179,"_145":180,"_182":183,"_184":185,"_187":188},"defaultBranch","main","fullName","Azure-Samples/remote-mcp-functions-typescript","githubTreeId",406531,8688,"language","Bicep","remote-mcp-functions-typescript",{"_149":150,"_106":181},"meBrEgjQp4","recentStargazers",1,"spdxLicense",{"_31":186},"MIT License","stargazers",40,"glamaJson","inspectable",false,"npmPackage","supportedPlatforms",[195,196,197],"WINDOWS","LINUX","MACOS","scores",{"_200":201,"_202":-5,"_203":-5},"license",100,"quality","security","simpleIcon","toolCount",0,"9khl8qvtum","updatedAt","2025-10-02T16:30:27.060126Z","meta",[212,215,217,224,227,229,232,235,238,241,244,246,249,252,254,256,258,260,262],{"_213":214},"title","Remote MCP Server | Glama",{"_216":40,"_31":29},"content",{"_218":219,"_220":221,"_222":223},"href","https://glama.ai/mcp/servers/@Azure-Samples/remote-mcp-functions-typescript","rel","canonical","tagName","link",{"_216":41,"_225":226},"property","og:title",{"_216":40,"_225":228},"og:description",{"_216":230,"_225":231},"https://glama.ai/logo.png","og:logo",{"_216":233,"_225":234},"https://glama.ai/mcp/servers/@Azure-Samples/remote-mcp-functions-typescript/og-image","og:image",{"_216":236,"_225":237},"630","og:image:height",{"_216":239,"_225":240},"1200","og:image:width",{"_216":242,"_225":243},"website","og:type",{"_216":219,"_225":245},"og:url",{"_216":247,"_225":248},"Glama – MCP Hosting Platform","og:site_name",{"_216":250,"_31":251},"summary_large_image","twitter:card",{"_216":247,"_31":253},"twitter:title",{"_216":40,"_31":255},"twitter:description",{"_216":233,"_31":257},"twitter:image",{"_216":236,"_31":259},"twitter:image:height",{"_216":239,"_31":261},"twitter:image:width",{"_263":264},"script:ld+json",{"_265":266,"_267":268},"@context","http://schema.org","@graph",[269],{"_265":270,"_271":272,"_273":274},"https://schema.org","@type","BreadcrumbList","itemListElement",[275,281,285,289],{"_271":276,"_277":278,"_31":279,"_280":183},"ListItem","item","/","Glama","position",{"_271":276,"_277":282,"_31":283,"_280":284},"/mcp","MCP",2,{"_271":276,"_277":286,"_31":287,"_280":288},"/mcp/servers","Servers",3,{"_271":276,"_277":290,"_31":41,"_280":291},"/mcp/servers/@Azure-Samples/remote-mcp-functions-typescript",4,"visitorIsMaintainer","routes/_public/mcp/servers/~namespace/~slug/_pages/_explorer/blob/~commit/~/_route",{"_295":296,"_27":302,"_330":331,"_210":343,"_340":367,"_368":369},"blob",{"_216":297,"_298":299,"_300":301},"\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n","fileName","architecture-diagram.drawio","size",4892,{"_83":84,"_13":303,"_92":40,"_93":40,"_16":94,"_95":304,"_31":41,"_143":317,"_151":-5,"_152":-5,"_153":320,"_198":329,"_204":-5,"_103":179,"_205":206,"_106":207,"_208":209},[86,87,88,89,90,91],[305,307,309,311,313,315],{"_98":306,"_29":105,"_106":107},{"_100":101,"_31":102,"_103":104},{"_98":308,"_29":113,"_106":114},{"_100":110,"_31":111,"_103":112},{"_98":310,"_29":120,"_106":121},{"_100":117,"_31":118,"_103":119},{"_98":312,"_29":127,"_106":128},{"_100":124,"_31":125,"_103":126},{"_98":314,"_29":134,"_106":135},{"_100":131,"_31":132,"_103":133},{"_98":316,"_29":141,"_106":142},{"_100":138,"_31":139,"_103":140},{"_145":318,"_103":150},{"_147":319},{"_149":150},{"_155":321,"_163":324,"_168":325,"_189":-5,"_190":191,"_192":-5,"_193":328},{"_157":322,"_159":323,"_161":162},{},[],{"_33":165,"_166":167},{"_170":171,"_172":173,"_174":175,"_16":176,"_177":178,"_31":179,"_145":326,"_182":183,"_184":327,"_187":188},{"_149":150,"_106":181},{"_31":186},[195,196,197],{"_200":201,"_202":-5,"_203":-5},"member",{"_332":333,"_336":337,"_338":299,"_174":175,"_339":191,"_340":299,"_341":342,"_300":301,"_161":295},"commit",{"_334":335},"sha","803b33b2e3e15e9254073d32e119bc54a4eee1ce","contentStoredObjectPointerId",365647,"fullPath","indexable","path","safeToDisplayAsText",true,[344,346,348,350,352,353,354,356,357,358,359,360,361,362,363,364,365,366],{"_213":345},"architecture-diagram.drawio | Remote MCP Server | Glama",{"_216":347,"_31":29},"Enterprise-grade security, privacy, and features like agents, MCP, prompt templates, and more.",{"_218":349,"_220":221,"_222":223},"https://glama.ai/mcp/servers/@Azure-Samples/remote-mcp-functions-typescript/blob/803b33b2e3e15e9254073d32e119bc54a4eee1ce/",{"_216":351,"_225":226},"architecture-diagram.drawio | Remote MCP Server",{"_216":347,"_225":228},{"_216":230,"_225":231},{"_216":355,"_225":234},"https://glama.ai/generated-images/og?title=architecture-diagram.drawio+%7C+Remote+MCP+Server",{"_216":236,"_225":237},{"_216":239,"_225":240},{"_216":242,"_225":243},{"_216":349,"_225":245},{"_216":247,"_225":248},{"_216":250,"_31":251},{"_216":247,"_31":253},{"_216":347,"_31":255},{"_216":355,"_31":257},{"_216":236,"_31":259},{"_216":239,"_31":261},"","treeMembers",[370,374,377,380,383,386,389,391,394,397,400,403,406,409,412,415,418,421],{"_332":371,"_338":372,"_340":372,"_161":373},{"_334":335},".devcontainer","tree",{"_332":375,"_338":376,"_340":376,"_161":373},{"_334":335},".github",{"_332":378,"_338":379,"_340":379,"_161":373},{"_334":335},"images",{"_332":381,"_338":382,"_340":382,"_161":373},{"_334":335},"infra",{"_332":384,"_338":385,"_340":385,"_161":373},{"_334":335},"src",{"_332":387,"_338":388,"_340":388,"_161":373},{"_334":335},".vscode",{"_332":390,"_338":299,"_340":299,"_161":295},{"_334":335},{"_332":392,"_338":393,"_340":393,"_161":295},{"_334":335},"architecture-diagram.png",{"_332":395,"_338":396,"_340":396,"_161":295},{"_334":335},"azure.yaml",{"_332":398,"_338":399,"_340":399,"_161":295},{"_334":335},".funcignore",{"_332":401,"_338":402,"_340":402,"_161":295},{"_334":335},".gitignore",{"_332":404,"_338":405,"_340":405,"_161":295},{"_334":335},"host.json",{"_332":407,"_338":408,"_340":408,"_161":295},{"_334":335},"LICENSE.md",{"_332":410,"_338":411,"_340":411,"_161":295},{"_334":335},"local.settings.json",{"_332":413,"_338":414,"_340":414,"_161":295},{"_334":335},"package.json",{"_332":416,"_338":417,"_340":417,"_161":295},{"_334":335},"package-lock.json",{"_332":419,"_338":420,"_340":420,"_161":295},{"_334":335},"README.md",{"_332":422,"_338":423,"_340":423,"_161":295},{"_334":335},"tsconfig.json","actionData","errors"]

MCP directory API