Azure-Samples · hallvictoria · May 1, 2026
diff --git a/src/FunctionsMcpResources/README.md b/src/FunctionsMcpResources/README.md
@@ -0,0 +1,219 @@
+# FunctionsMcpResources — MCP Resource Templates on Azure Functions (Python)
+
+This project is a Python Azure Function app that exposes MCP (Model Context Protocol) resource templates as a remote MCP server. Resource templates allow MCP clients to discover and read structured data through URI-based patterns.
+
+> **Note:** MCP tools are in the [FunctionsMcpTool](../FunctionsMcpTool) project, and prompts are in the [FunctionsMcpPrompts](../FunctionsMcpPrompts) project.
+
+## Resources included
+
+| Resource | URI | Description |
+|----------|-----|-------------|
+| `Snippet` | `snippet://{Name}` | Resource template that reads a code snippet by name from blob storage. Clients discover it via `resources/templates/list` and substitute the `Name` parameter. |
+| `ServerInfo` | `info://server` | Static resource that returns server name, version, runtime, and timestamp. |
+
+## Key concepts
+
+- **Resource templates** have URI parameters (e.g., `{Name}`) that clients substitute at runtime — they're like parameterized endpoints.
+- **Static resources** have fixed URIs and return the same structure every call.
+- **Resource metadata** (like cache TTL) can be passed in the `metadata` parameter of the `@app.mcp_resource_trigger` decorator.
+
+## Prerequisites
+
+- [Python 3.13+](https://www.python.org/downloads/)
+- [Azure Functions Core Tools](https://learn.microsoft.com/azure/azure-functions/functions-run-local?pivots=programming-language-python#install-the-azure-functions-core-tools) >= `4.5.0`
+- [Docker](https://www.docker.com/) (for the Azurite storage emulator — needed by the snippet resource template)
+
+## Run locally
+
+### 1. Start Azurite (required for the snippet resource which uses blob storage)
+
+```bash
+docker run -d -p 10000:10000 -p 10001:10001 -p 10002:10002 \
+  mcr.microsoft.com/azure-storage/azurite
+```
+
+### 2. Upload sample snippets to Azurite
+
+Once Azurite is running, you need to upload the sample snippet files to blob storage. You can use [Azure Storage Explorer](https://azure.microsoft.com/features/storage-explorer/) or the Azure CLI:
+
+```bash
+# Using Azure CLI with Azurite connection string
+az storage container create --name snippets \
+  --connection-string "DefaultEndpointsProtocol=http;AccountName=devstoreaccount1;AccountKey=Eby8vdM02xNOcqFlqUwJPLlmEtlCDXJ1OUzFT50uSRZ6IFsuFq2UVErCz4I6tq/K1SZFPTOtr/KBHBeksoGMGw==;BlobEndpoint=http://127.0.0.1:10000/devstoreaccount1;"
+
+# Upload sample snippets
+az storage blob upload-batch --source ../../__queuestorage__/snippets \
+  --destination snippets \
+  --connection-string "DefaultEndpointsProtocol=http;AccountName=devstoreaccount1;AccountKey=Eby8vdM02xNOcqFlqUwJPLlmEtlCDXJ1OUzFT50uSRZ6IFsuFq2UVErCz4I6tq/K1SZFPTOtr/KBHBeksoGMGw==;BlobEndpoint=http://127.0.0.1:10000/devstoreaccount1;"
+```
+
+### 3. Start the Functions host
+
+From this directory (`src/FunctionsMcpResources`), start the Functions host:
+
+```bash
+func start
+```
+
+The MCP endpoint will be available at `http://localhost:7073/runtime/webhooks/mcp`.
+
+## Deploy to Azure
+
+From the repository root, use `azd` to deploy:
+
+```bash
+azd env set DEPLOY_SERVICE resources
+azd provision
+azd deploy --service resources
+```
+
+## Examining the code
+
+Resources are defined in [function_app.py](function_app.py). Each resource is a Python function with an `@app.mcp_resource_trigger` decorator:
+
+### Resource Template (with URI parameter)
+
+```python
+@app.mcp_resource_trigger(
+    arg_name="context",
+    uri="snippet://{Name}",
+    resource_name="Snippet",
+    description="Reads a code snippet by name from blob storage.",
+    mime_type="application/json"
+)
+@app.blob_input(
+    arg_name="snippet_content",
+    path="snippets/{mcpresourceargs.Name}.json",
+    connection="AzureWebJobsStorage"
+)
+def get_snippet_resource(context, snippet_content: Optional[bytes]) -> str:
+    # The {mcpresourceargs.Name} binding expression automatically extracts
+    # the Name parameter from the resource URI and passes it to the blob binding
+    if snippet_content is None:
+        return json.dumps({"error": "Snippet not found"})
+    return snippet_content.decode('utf-8')
+```
+
+The `{mcpresourceargs.Name}` binding expression automatically extracts the `Name` parameter from the resource URI and passes it to the blob input binding.
+
+### Static Resource (no parameters)
+
+```python
+@app.mcp_resource_trigger(
+    arg_name="context",
+    uri="info://server",
+    resource_name="ServerInfo",
+    description="Returns information about the MCP server.",
+    mime_type="application/json",
+    metadata=json.dumps({"cache": {"ttlSeconds": 60}})
+)
+def get_server_info(context) -> str:
+    server_info = {
+        "name": "FunctionsMcpResources",
+        "version": "1.0.0",
+        "runtime": f"Python {platform.python_version()}",
+        "timestamp": datetime.now(timezone.utc).isoformat()
+    }
+    return json.dumps(server_info)
+```
+
+## Testing the resources
+
+### Using MCP Inspector
+
+Install and use the [MCP Inspector](https://github.com/modelcontextprotocol/inspector) to test your resources:
+
+```bash
+npx @modelcontextprotocol/inspector http://localhost:7073/runtime/webhooks/mcp
+```
+
+### Using curl
+
+Test the ServerInfo static resource:
+
+```bash
+curl -X POST http://localhost:7073/runtime/webhooks/mcp \
+  -H "Content-Type: application/json" \
+  -d '{
+    "jsonrpc": "2.0",
+    "id": 1,
+    "method": "resources/read",
+    "params": {
+      "uri": "info://server"
+    }
+  }'
+```
+
+List available resource templates:
+
+```bash
+curl -X POST http://localhost:7073/runtime/webhooks/mcp \
+  -H "Content-Type: application/json" \
+  -d '{
+    "jsonrpc": "2.0",
+    "id": 1,
+    "method": "resources/templates/list"
+  }'
+```
+
+Read a specific snippet:
+
+```bash
+curl -X POST http://localhost:7073/runtime/webhooks/mcp \
+  -H "Content-Type: application/json" \
+  -d '{
+    "jsonrpc": "2.0",
+    "id": 1,
+    "method": "resources/read",
+    "params": {
+      "uri": "snippet://HelloWorld"
+    }
+  }'
+```
+
+## Architecture
+
+```
+┌─────────────────┐
+│   MCP Client    │
+│  (e.g., Agent)  │
+└────────┬────────┘
+         │
+         │ HTTP
+         │
+┌────────▼────────────────────────┐
+│   Azure Functions (Python)      │
+│  ┌──────────────────────────┐   │
+│  │  get_snippet_resource    │   │
+│  │  (Resource Template)     │   │
+│  └──────────┬───────────────┘   │
+│             │                   │
+│             │ Blob Binding      │
+│             │                   │
+│  ┌──────────▼───────────────┐   │
+│  │   Azure Blob Storage     │   │
+│  │   (snippets container)   │   │
+│  └──────────────────────────┘   │
+│                                 │
+│  ┌──────────────────────────┐   │
+│  │   get_server_info        │   │
+│  │   (Static Resource)      │   │
+│  └──────────────────────────┘   │
+└─────────────────────────────────┘
+```
+
+## Sample snippets
+
+Three sample snippets are included in `__queuestorage__/snippets/`:
+
+1. **HelloWorld.json** - A simple Hello World function
+2. **QuickSort.json** - QuickSort algorithm implementation
+3. **FibonacciSequence.json** - Fibonacci sequence generator
+
+You can add more snippets by creating JSON files in the same format and uploading them to the blob storage container.
+
+## Related documentation
+
+- [Model Context Protocol Specification](https://spec.modelcontextprotocol.io/)
+- [Azure Functions Python Developer Guide](https://learn.microsoft.com/azure/azure-functions/functions-reference-python)
+- [Azure Functions Blob Storage Bindings](https://learn.microsoft.com/azure/azure-functions/functions-bindings-storage-blob)
diff --git a/src/FunctionsMcpResources/function_app.py b/src/FunctionsMcpResources/function_app.py
@@ -0,0 +1,151 @@
+"""
+FunctionsMcpResources - MCP Resource Templates on Azure Functions (Python)
+
+This module demonstrates both resource templates and static resources using Azure Functions.
+- Resource templates have URI parameters (e.g., {Name}) that clients substitute at runtime
+- Static resources have fixed URIs and return consistent data structures
+"""
+
+import json
+import logging
+import os
+import platform
+import re
+from datetime import datetime, timezone
+
+import azure.functions as func
+from azure.storage.blob import BlobServiceClient
+
+app = func.FunctionApp(http_auth_level=func.AuthLevel.FUNCTION)
+
+# ============================================================================
+# Resource URI and Metadata Constants
+# ============================================================================
+
+# Snippet Resource Template (has URI parameter {Name})
+SNIPPET_RESOURCE_URI = "snippet://{Name}"
+SNIPPET_RESOURCE_NAME = "Snippet"
+SNIPPET_RESOURCE_DESCRIPTION = "Reads a code snippet by name from blob storage."
+SNIPPET_MIME_TYPE = "application/json"
+
+# Server Info Static Resource (fixed URI, no parameters)
+SERVER_INFO_RESOURCE_URI = "info://server"
+SERVER_INFO_RESOURCE_NAME = "ServerInfo"
+SERVER_INFO_RESOURCE_DESCRIPTION = "Returns information about the MCP server, including version and runtime."
+SERVER_INFO_MIME_TYPE = "application/json"
+
+# Metadata for ServerInfo resource (cache TTL)
+SERVER_INFO_METADATA = json.dumps({"cache": {"ttlSeconds": 60}})
+
+# ============================================================================
+# Resource Template: Snippet
+# ============================================================================
+
+@app.mcp_resource_trigger(
+    arg_name="context",
+    uri=SNIPPET_RESOURCE_URI,
+    resource_name=SNIPPET_RESOURCE_NAME,
+    description=SNIPPET_RESOURCE_DESCRIPTION,
+    mime_type=SNIPPET_MIME_TYPE
+)
+def get_snippet_resource(context) -> str:
+    """
+    Resource template that exposes snippets by name.
+
+    The {Name} parameter in the URI makes this a resource template rather than
+    a static resource — clients can discover it via resources/templates/list
+    and read specific snippets by substituting the Name parameter.
+
+    This implementation manually extracts the Name parameter from the URI
+    and uses the Azure Blob Storage SDK to read the corresponding blob.
+
+    Args:
+        context: MCP resource invocation context
+
+    Returns:
+        JSON string containing the snippet content or an error message
+    """
+    logging.info(f"Snippet resource template invoked: {context.uri}")
+
+    try:
+        # Extract the Name parameter from the URI (e.g., "snippet://HelloWorld" -> "HelloWorld")
+        # The URI pattern is "snippet://{Name}"
+        match = re.match(r"snippet://(.+)", context.uri)
+        if not match:
+            error_response = {
+                "error": "Invalid URI",
+                "message": f"URI does not match expected pattern 'snippet://{{Name}}'"
+            }
+            return json.dumps(error_response)
+
+        snippet_name = match.group(1)
+        logging.info(f"Extracted snippet name: {snippet_name}")
+
+        # Get the blob storage connection string
+        connection_string = os.environ.get("AzureWebJobsStorage")
+        if not connection_string:
+            error_response = {
+                "error": "Configuration error",
+                "message": "AzureWebJobsStorage connection string not found"
+            }
+            return json.dumps(error_response)
+
+        # Create blob service client and read the blob
+        blob_service_client = BlobServiceClient.from_connection_string(connection_string)
+        container_client = blob_service_client.get_container_client("snippets")
+        blob_client = container_client.get_blob_client(f"{snippet_name}.json")
+
+        # Download the blob content
+        blob_data = blob_client.download_blob()
+        snippet_content = blob_data.readall().decode('utf-8')
+
+        return snippet_content
+
+    except Exception as e:
+        logging.error(f"Error reading snippet: {e}")
+        error_response = {
+            "error": "Snippet not found",
+            "message": f"No snippet found for the requested name. Error: {str(e)}"
+        }
+        return json.dumps(error_response)
+
+# ============================================================================
+# Static Resource: Server Info
+# ============================================================================
+
+@app.mcp_resource_trigger(
+    arg_name="context",
+    uri=SERVER_INFO_RESOURCE_URI,
+    resource_name=SERVER_INFO_RESOURCE_NAME,
+    description=SERVER_INFO_RESOURCE_DESCRIPTION,
+    mime_type=SERVER_INFO_MIME_TYPE,
+    metadata=SERVER_INFO_METADATA
+)
+def get_server_info(context) -> str:
+    """
+    Static resource (no URI parameters) that returns server information.
+
+    Demonstrates the difference between a static resource and a resource template.
+    This resource has a fixed URI with no parameters and returns dynamic server
+    metadata each time it's invoked.
+
+    The cache metadata (ttlSeconds: 60) hints to clients that they can cache
+    this resource for 60 seconds.
+
+    Args:
+        context: MCP resource invocation context
+
+    Returns:
+        JSON string containing server information
+    """
+    logging.info("Server info resource invoked.")
+
+    server_info = {
+        "name": "FunctionsMcpResources",
+        "version": "1.0.0",
+        "runtime": f"Python {platform.python_version()}",
+        "platform": platform.platform(),
+        "timestamp": datetime.now(timezone.utc).isoformat()
+    }
+
+    return json.dumps(server_info, indent=2)
diff --git a/src/FunctionsMcpResources/host.json b/src/FunctionsMcpResources/host.json
@@ -0,0 +1,23 @@
+{
+  "version": "2.0",
+  "extensions": {
+    "mcp": {
+      "system": {
+        "webhookAuthorizationLevel": "Anonymous"
+      }
+    }
+  },
+  "logging": {
+    "applicationInsights": {
+      "samplingSettings": {
+        "isEnabled": true,
+        "excludedTypes": "Request"
+      },
+      "enableLiveMetricsFilters": true
+    }
+  },
+  "extensionBundle": {
+    "id": "Microsoft.Azure.Functions.ExtensionBundle",
+    "version": "[4.*, 5.0.0)"
+  }
+}