☁️ 🚀 Build and Deploy a remote MCP server to Cloud Run in 5 minutes

Anthropic's Model Context Protocol (MCP) is a protocol that helps you connect prompts, resources and tools (context) to an LLM.

MCP has become increasingly popular with the rise of code assistants and AI-powered IDEs for software development, as well as agentic applications that need to connect to external tools and resources.

MCP follows a client-server architecture, and for the longest time only supported running the server locally using the stdio transport.

MCP now supports two additional transports: sse and streamable-http, which allow for running MCP servers remotely. You can read more about these transports in the official MCP docs.

Benefits of running an MCP server remotely

Running an MCP server remotely on Cloud Run can provide several benefits:

• 📈 Scalability: Cloud Run is built to rapidly scale out to handle all incoming requests. Cloud Run will scale your MCP server automatically based on demand.
• 👥 Centralized server: You can share your MCP server with team members in one centralized location and they can connect to it from their local machines instead of running their own servers locally. If a change is made to the MCP server, all team members will benefit from it.
• 🔐 Security: Cloud Run provides an easy way to force authenticated requests. This allows only secure connections to your MCP server, preventing unauthorized access.

🔢 Math MCP Server Example

We will use FastMCP to create a simple math MCP server that has two tools: add and subtract. FastMCP provides a fast, Pythonic way to build MCP servers and clients.

🛠️ Tools

The MCP server has the following two tools:

The add tool will take two numbers and return their sum.

https://github.com/jackwotherspoon/mcp-on-cloudrun/blob/87960732c42f597c60d6d6aba3a2634aa98f4361/server.py#L25-L37

The subtract tool will take two numbers and return their difference.

https://github.com/jackwotherspoon/mcp-on-cloudrun/blob/87960732c42f597c60d6d6aba3a2634aa98f4361/server.py#L39-L51

Transport

We will use the streamable-http transport for this example as it is the recommended transport for remote servers as of the 2025-03-26 spec, but you can also still use sse if you prefer as it is backwards compatible.

If you want to use sse you will need to update the last line of server.py to use transport="sse".

https://github.com/jackwotherspoon/mcp-on-cloudrun/blob/a5a386c8e744cfa9b05eb5f021953a56b33bf79f/server.py#L55-L56

Prerequisites

• Python 3.10+ (FastMCP requires 3.10+)
• Google Cloud SDK (gcloud)
• Git (for cloning the repository)

Installation

Clone the repository

git clone https://github.com/jackwotherspoon/mcp-on-cloudrun.git
cd mcp-on-cloudrun

Set your Google Cloud credentials and project.

gcloud auth login
export PROJECT_ID=<your-project-id>
gcloud config set project $PROJECT_ID

Deploying to Cloud Run

Now let's deploy a simple MCP server to Cloud Run. You can deploy directly from source or using a container image.

For both options we will use the --no-allow-unauthenticated flag to require authentication.

This is important for security reasons. If you don't require authentication, anyone can call your MCP server and potentially cause damage to your system.

gcloud run deploy mcp-server --no-allow-unauthenticated --region=us-central1 --source .

gcloud artifacts repositories create remote-mcp-servers \
  --repository-format=docker \
  --location=us-central1 \
  --description="Repository for remote MCP servers" \
  --project=$PROJECT_ID

gcloud builds submit --region=us-central1 --tag us-central1-docker.pkg.dev/$PROJECT_ID/remote-mcp-servers/mcp-server:latest

gcloud run deploy mcp-server \
  --image us-central1-docker.pkg.dev/$PROJECT_ID/remote-mcp-servers/mcp-server:latest \
  --region=us-central1 \
  --no-allow-unauthenticated

If your service has successfully deployed you will see a message like the following:

Service [mcp-server] revision [mcp-server-12345-abc] has been deployed and is serving 100 percent of traffic.

Authenticating MCP Clients

Since we specified --no-allow-unauthenticated to require authentication, any MCP client connecting to our remote MCP server will need to authenticate.

The official docs for Host MCP servers on Cloud Run provides more information on this topic depending on where you are running your MCP client.

For this example we will run the Cloud Run proxy to create an authenticated tunnel to our remote MCP server on our local machines.

By default, the URL of Cloud Run services requires all requests to be authorized with the Cloud Run Invoker (roles/run.invoker) IAM role. This IAM policy binding ensures that a strong security mechanism is used to authenticate your local MCP client.

You should make sure that you or any team members trying to access the remote MCP server have the roles/run.invoker IAM role bound to their Google Cloud account.

gcloud run services proxy mcp-server --region=us-central1

You should see the following output:

Proxying to Cloud Run service [mcp-server] in project [<YOUR_PROJECT_ID>] region [us-central1]
http://127.0.0.1:8080 proxies to https://mcp-server-abcdefgh-uc.a.run.app

All traffic to http://127.0.0.1:8080 will now be authenticated and forwarded to our remote MCP server.

Testing our remote MCP server

Let's test and connect to our remote MCP server using the test_server.py test script. It uses the FastMCP client to connect to http://127.0.0.1:8080/mcp (note the /mcp at the end as we are using the streamable-http transport) and call the add and subtract tools.

In a new terminal run:

uv run test_server.py

You should see the following output:

>>> 🛠️  Tool found: add
>>> 🛠️  Tool found: subtract
>>> 🪛  Calling add tool for 1 + 2
<<< ✅ Result: 3
>>> 🪛  Calling subtract tool for 10 - 3
<<< ✅ Result: 7

You have done it! You have successfully deployed a remote MCP server to Cloud Run and tested it using the FastMCP client.