{"id":53,"date":"2026-09-10T14:40:40","date_gmt":"2026-09-10T14:40:40","guid":{"rendered":"https:\/\/roboticsmaestro.com\/ai\/?p=53"},"modified":"2026-09-10T16:39:55","modified_gmt":"2026-09-10T16:39:55","slug":"instructions-to-setup-mcp-server-with-function-app-for-microsoft-foundry-agent-tools-with-app-insights","status":"publish","type":"post","link":"https:\/\/roboticsmaestro.com\/ai\/blog\/2026\/09\/10\/instructions-to-setup-mcp-server-with-function-app-for-microsoft-foundry-agent-tools-with-app-insights\/","title":{"rendered":"Instructions to setup mcp server with function app for microsoft foundry agent tools with app insights"},"content":{"rendered":"\n

<\/p>\n\n\n\n

walk through this in a clear, step-by-step<\/strong> way so you can set up a new MCP (Microsoft Foundry Custom Plugin)<\/strong> using an Azure Function App<\/strong>, integrate it with Microsoft Foundry Tools<\/strong>, and enable Application Insights<\/strong> for monitoring.<\/p>\n\n\n\n

I\u2019ll break it into three main phases<\/strong>:<\/p>\n\n\n\n

    \n
  1. Azure Function App Setup<\/strong><\/li>\n\n\n\n
  2. MCP (Custom Tool) Registration in Foundry<\/strong><\/li>\n\n\n\n
  3. Application Insights Integration<\/strong><\/li>\n<\/ol>\n\n\n\n

    1. Create and Configure the Azure Function App<\/strong><\/h2>\n\n\n\n

    Step 1 \u2014 Create a Resource Group<\/strong><\/p>\n\n\n\n

    az group create --name my-foundry-mcp-rg --location eastus
    <\/code><\/pre>\n\n\n\n
    Step 2 \u2014 Create a Storage Account<\/strong><\/p>\n\n\n\n
    BashCopy codeaz storage account create \\\n  --name mystorageacctmcp \\\n  --location eastus \\\n  --resource-group my-foundry-mcp-rg \\\n  --sku Standard_LRS\n<\/code><\/pre>\n\n\n\n
    Step 3 \u2014 Create the Function App<\/strong><\/p>\n\n\n\n
    az functionapp create \\
      --resource-group my-foundry-mcp-rg \\
      --consumption-plan-location eastus \\
      --runtime python \\
      --runtime-version 3.10 \\
      --functions-version 4 \\
      --name my-foundry-mcp-func \\
      --storage-account mystorageacctmcp
    <\/code><\/pre>\n\n\n\n
    2. Build the MCP Logic in the Function App<\/strong><\/h2>\n\n\n\n
    Example: HttpTrigger<\/code> Function for File Retrieval<\/strong><\/p>\n\n\n\n
    PythonCopy code

    import logging
    import azure.functions as func
    from azure.storage.blob import BlobServiceClient
    import json

    def main(req: func.HttpRequest) -> func.HttpResponse:
        logging.info(\"MCP Function triggered.\")

        source_type = req.params.get('sourceType')
        if not source_type:
            return func.HttpResponse(\"Missing 'sourceType' parameter\", status_code=400)

        results = []

        if source_type.lower() == \"blob\":
            blob_service_client = BlobServiceClient.from_connection_string(
                \"<BLOB_CONNECTION_STRING>\"
            )
            container_client = blob_service_client.get_container_client(\"my-container\")
            for blob in container_client.list_blobs():
                results.append({
                    \"sourceType\": \"Blob\",
                    \"fileName\": blob.name.split(\"\/\")[-1],
                    \"filePath\": blob.name
                })

        elif source_type.lower() == \"sharepoint\":
            # Placeholder for SharePoint retrieval logic
            results.append({
                \"sourceType\": \"SharePoint\",
                \"fileName\": \"example.docx\",
                \"filePath\": \"\/Shared Documents\/example.docx\"
            })

        return func.HttpResponse(
            json.dumps(results),
            mimetype=\"application\/json\",
            status_code=200
        )
    <\/code><\/pre>\n\n\n\n
    Deploy this function:<\/p>\n\n\n\n
    func azure functionapp publish my-foundry-mcp-func
    <\/code><\/pre>\n\n\n\n
    3. Enable Application Insights<\/strong><\/h2>\n\n\n\n
    Step 1 \u2014 Create Application Insights<\/strong><\/p>\n\n\n\n
    BashCopy codeaz monitor app-insights component create \\\n  --app my-foundry-mcp-ai \\\n  --location eastus \\\n  --resource-group my-foundry-mcp-rg\n<\/code><\/pre>\n\n\n\n
    Step 2 \u2014 Link to Function App<\/strong><\/p>\n\n\n\n
    az functionapp update \\
      --name my-foundry-mcp-func \\
      --resource-group my-foundry-mcp-rg \\
      --set applicationInsights.key=<YOUR_APP_INSIGHTS_KEY>
    <\/code><\/pre>\n\n\n\n
    Step 3 \u2014 Add Telemetry in Code<\/strong><\/p>\n\n\n\n
    from opencensus.ext.azure.log_exporter import AzureLogHandler

    logger = logging.getLogger(__name__)
    logger.addHandler(AzureLogHandler(
        connection_string='InstrumentationKey=<YOUR_APP_INSIGHTS_KEY>')
    )

    logger.warning(\"Custom MCP function executed.\")
    <\/code><\/pre>\n\n\n\n
    4. Register the MCP in Microsoft Foundry Tools<\/strong><\/h2>\n\n\n\n
    Step 1 \u2014 Go to Foundry Agent Service<\/strong><\/p>\n\n\n\n
      \n
    • Navigate to\u00a0Foundry \u2192 Agent Tools \u2192 Add Tool<\/strong>.<\/li>\n\n\n\n
    • Select\u00a0Custom HTTP Tool<\/strong>.<\/li>\n<\/ul>\n\n\n\n
      Step 2 \u2014 Configure Tool<\/strong><\/p>\n\n\n\n
        \n
      • Name<\/strong>:\u00a0RetrieveFilesTool<\/code><\/li>\n\n\n\n
      • Endpoint URL<\/strong>:\u00a0https:\/\/my-foundry-mcp-func.azurewebsites.net\/api\/<function-name><\/code><\/li>\n\n\n\n
      • Method<\/strong>:\u00a0GET<\/code><\/li>\n\n\n\n
      • Parameters<\/strong>:\n
          \n
        • sourceType<\/code>\u00a0(string, required:\u00a0\"Blob\"<\/code>\u00a0or\u00a0\"SharePoint\"<\/code>)<\/li>\n<\/ul>\n<\/li>\n<\/ul>\n\n\n\n
          Step 3 \u2014 Test in Foundry<\/strong><\/p>\n\n\n\n
            \n
          • Run a test query:
            “Retrieve all file paths from Blob storage”<\/em><\/li>\n\n\n\n
          • Foundry will call your Function App MCP and return JSON results.<\/li>\n<\/ul>\n\n\n\n
            ✅ At this point<\/strong> you have:<\/p>\n\n\n\n
              \n
            • An\u00a0Azure Function App<\/strong>\u00a0acting as your MCP backend.<\/li>\n\n\n\n
            • Application Insights<\/strong>\u00a0capturing telemetry.<\/li>\n\n\n\n
            • The MCP registered as a\u00a0Foundry Tool<\/strong>\u00a0for your agents<\/li>\n<\/ul>\n\n\n\n
              <\/p>\n\n\n\n
              proceed<\/strong> by taking what we\u2019ve built and making it production-ready<\/strong> so your MCP + Azure Function App + Foundry integration is reliable, secure, and observable.<\/p>\n\n\n\n
              I\u2019ll walk you through the next steps<\/strong>: securing the function, improving telemetry, and validating the MCP inside Foundry.<\/p>\n\n\n\n
              5. Secure the Azure Function App<\/strong><\/h2>\n\n\n\n
              By default, Azure Functions can be called publicly. For a production MCP, you\u2019ll want to restrict access<\/strong> so only Foundry can call it.<\/p>\n\n\n\n
              Option A \u2014 Function Key Authentication<\/strong><\/p>\n\n\n\n
                \n
              1. Get the function key:BashCopy codeaz functionapp function keys list \\ --name my-foundry-mcp-func \\ --resource-group my-foundry-mcp-rg \\ --function-name <function-name><\/code><\/li>\n\n\n\n
              2. In Foundry Tool configuration, add a header:Copy codex-functions-key: <FUNCTION_KEY><\/code><\/li>\n<\/ol>\n\n\n\n
                Option B \u2014 Azure AD Authentication (Recommended)<\/strong><\/p>\n\n\n\n
                  \n
                1. Enable\u00a0Azure AD Authentication<\/strong>\u00a0in the Function App:BashCopy codeaz webapp auth microsoft update \\ --resource-group my-foundry-mcp-rg \\ --name my-foundry-mcp-func \\ --enabled true<\/code><\/li>\n\n\n\n
                2. Register Foundry as an\u00a0App Registration<\/strong>\u00a0in Azure AD and grant it API permissions to call your function.<\/li>\n<\/ol>\n\n\n\n
                  6. Enhance Application Insights Telemetry<\/strong><\/h2>\n\n\n\n
                  We can make telemetry more useful by tracking:<\/p>\n\n\n\n
                    \n
                  • Request parameters<\/strong>\u00a0(without sensitive data)<\/li>\n\n\n\n
                  • Execution time<\/strong><\/li>\n\n\n\n
                  • Error details<\/strong><\/li>\n<\/ul>\n\n\n\n
                    Example:<\/strong><\/p>\n\n\n\n
                    PythonCopy codeimport time\nfrom opencensus.ext.azure.log_exporter import AzureLogHandler\n\nlogger = logging.getLogger(__name__)\nlogger.addHandler(AzureLogHandler(\n    connection_string='InstrumentationKey=<YOUR_APP_INSIGHTS_KEY>')\n)\n\ndef main(req: func.HttpRequest) -> func.HttpResponse:\n    start_time = time.time()\n    try:\n        source_type = req.params.get('sourceType')\n        logger.info(f\"Request received for sourceType={source_type}\")\n\n        # ... existing logic ...\n\n        duration = time.time() - start_time\n        logger.info(f\"Execution completed in {duration:.2f}s\")\n        return func.HttpResponse(json.dumps(results), mimetype=\"application\/json\")\n\n    except Exception as e:\n        logger.error(f\"Error occurred: {str(e)}\", exc_info=True)\n        return func.HttpResponse(\"Internal Server Error\", status_code=500)\n<\/code><\/pre>\n\n\n\n
                    7. Validate MCP in Foundry<\/strong><\/h2>\n\n\n\n
                    Step 1 \u2014 Test Tool in Foundry<\/strong><\/p>\n\n\n\n