FunctionsMcpResources

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 project, and prompts are in the 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+
• Azure Functions Core Tools >= 4.8.0
• Azure Developer CLI (azd) 1.23.x or above (for deployment)
• Docker (for the Azurite storage emulator — needed by the snippet resource template)

Prepare your local environment

An Azure Storage Emulator is needed because the snippet resource reads blobs from storage. Start Azurite:

docker run -d -p 10000:10000 -p 10001:10001 -p 10002:10002 \
    mcr.microsoft.com/azure-storage/azurite \
    azurite --skipApiVersionCheck --blobHost 0.0.0.0 --queueHost 0.0.0.0 --tableHost 0.0.0.0

If you use the Azurite VS Code extension instead, run Azurite: Start now.

Run locally

1. Install dependencies

Create and activate a virtual environment, then install dependencies:

python3 -m venv .venv
source .venv/bin/activate    # macOS/Linux
.venv\Scripts\activate       # Windows
pip install -r requirements.txt

2. Start the Functions host

From this directory (src/FunctionsMcpResources), start the Functions host:

func start

The MCP endpoint will be available at http://localhost:7071/runtime/webhooks/mcp.

3. Seed a snippet using the save_snippet tool

This project includes a save_snippet MCP tool that writes snippets to blob storage. Use it to seed data before reading resources. For example, from VS Code Chat or MCP Inspector:

"Save a snippet called HelloWorld with the content: print('Hello, World!')"

This creates a blob at snippets/HelloWorld.json which can then be read via the snippet://HelloWorld resource.

Connect to the MCP server

Option A: VS Code with GitHub Copilot

Open .vscode/mcp.json in the workspace root. Find the server called local-mcp-function and click Start. It points to:

http://localhost:7071/runtime/webhooks/mcp

MCP resources are attached as context in VS Code Chat (they aren't invoked like tools or prompts):

1. Open the Chat panel.
2. Click the + (Attach) button in the chat input.
3. Select MCP Resources.
4. Choose a resource:
   • ServerInfo — no parameters needed. Returns server name, version, runtime, and timestamp.
   • Snippet — enter a snippet name (e.g., HelloWorld). Reads the matching blob from storage.
5. The resource content is attached to the conversation as context for the model.

Option B: MCP Inspector

1. Install and run MCP Inspector:

   npx @modelcontextprotocol/inspector
   

2. Open the Inspector URL (e.g. http://0.0.0.0:5173/#resources).

3. Set the transport type to Streamable HTTP.

4. Set the URL to http://0.0.0.0:7071/runtime/webhooks/mcp and click Connect.

5. Click List Resources or List Resource Templates to browse available resources.

Deploy to Azure

Step 1: Sign in

az login
azd auth login

Step 2: Create an environment

azd env new <environment-name>

This also becomes the resource group name.

Step 3: Provision and deploy

By default, OAuth-based authentication is enabled using the built-in MCP auth feature with Microsoft Entra as the identity provider.

Configure VS Code as an allowed client application for Microsoft Entra:

azd env set PRE_AUTHORIZED_CLIENT_IDS aebc6443-996d-45c2-90f0-388ff96faa56

Optionally enable VNet isolation:

azd env set VNET_ENABLED true

Deploy the project. When prompted, pick your subscription and an Azure region.

azd up

Step 4: Connect to the remote MCP server

Open .vscode/mcp.json and click Start above remote-mcp-function. You'll be prompted for functionapp-name — find it in your azd command output or the .azure/<env>/.env file. Since authentication is enabled, you'll also be prompted to sign in with Microsoft.

Tip: Click More... → Show Output above the server name to see request/response details.

Redeploy and clean up

• Redeploy: azd deploy
• Clean up all resources: azd down

Examining the code

Resources are defined in function_app.py. Each resource is a Python function with an @app.mcp_resource_trigger decorator:

Resource Template (with URI parameter)

@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:
    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)

@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)

Troubleshooting

Problem	Solution
Connection refused locally	Ensure Azurite is running (docker run -p 10000:10000 ...)
API version not supported by Azurite	Add --skipApiVersionCheck flag to the Azurite command, or pull the latest image
AttributeError: 'FunctionApp' object has no attribute 'mcp_resource_trigger'	Python 3.13 is required. Verify with python3 --version.
azd up provision succeeded but deploy failed	Transient error — run azd deploy again