How to call a python MCP tool hosted on Google Cloud Run

I have deployed a python script for mcp server in a docker container on Google Cloud Run. Below is a sample script

import asyncio
import logging
import os

from fastmcp import FastMCP 

logger = logging.getLogger(__name__)
logging.basicConfig(format="[%(levelname)s]: %(message)s", level=logging.INFO)

mcp = FastMCP("MCP Server on Cloud Run")

@mcp.tool()
def add(a: int, b: int) -> int:
    """Use this to add two numbers together.
    
    Args:
        a: The first number.
        b: The second number.
    
    Returns:
        The sum of the two numbers.
    """
    logger.info(f">>> Tool: 'add' called with numbers '{a}' and '{b}'")
    return a + b

if __name__ == "__main__":
    logger.info(f" MCP server started on port {os.getenv('PORT', 8080)}")
    # Could also use 'sse' transport, host="0.0.0.0" required for Cloud Run.
    asyncio.run(
        mcp.run_async(
            transport="streamable-http", 
            host="0.0.0.0", 
            port=os.getenv("PORT", 8080),
        )
    ) 

I have put this in docker and deployed the image to CloudRun and got the https endpoint for calling a streamable https request.

I have created a Service account with Cloud Run Invoker permission and generated a json key. But when I try to access the service from python I am getting 403 unauthorised error. I used the below code to try to call the mcp server.

import os
import json
import requests
import google.oauth2.id_token
import google.auth.transport.requests

def runCloudFunction():
    os.environ['GOOGLE_APPLICATION_CREDENTIALS'] = 'path\to\file.json'
    request = google.auth.transport.requests.Request()
    audience = 'https://cloud_run_service_url'
    TOKEN = google.oauth2.id_token.fetch_id_token(request, audience)
    print(TOKEN)
    r=requests.post(
        audience+'/mcp',
        headers={'Authorization':"Bearer "+TOKEN, 'Content-Type':'application/json'})
    print(r.status_code)

if __name__ == "__main__":
    runCloudFunction()    

The above code is printing the token but returning status 403 for the request to the service. I do not want to remove authentication from the service since that will make it insecure.So , I have selected the Require Authentication option.

I checked that public access is enabled for the Cloud Service in Networking Settings.

Will be really grateful if someone can let me know what I missed. I am not aware of the body to pass to the service to call a particular python function/mcp-tool. WIll be helpful if someone can guide on that as well. Thank you in advance.

The following works for me.

In reverse order:

export GOOGLE_APPLICATION_CREDENTIALS=${PWD}/tester.json
export CLOUD_RUN_URL=$(\
  gcloud run services describe ${NAME} \
  --region=${REGION} \
  --project=${PROJECT} \
  --format="value(status.url)")

uv run add.py 25 17 # add(25,17)

Yields:

Connected
[TextContent(type='text', text='42', annotations=None)]

With: pyproject.toml

[project]
name = "79685701"
version = "0.0.1"
description = "Stackoverflow: 79685701"
readme = "README.md"
requires-python = ">=3.13"
dependencies = [
    "fastmcp>=2.9.2",
    "google-auth>=2.40.3",
    "requests>=2.32.4"
]

main.py:

import asyncio
import os

from fastmcp import FastMCP, Context

mcp = FastMCP("MCP Server on Cloud Run")

@mcp.tool()
async def add(a: int, b: int, ctx: Context) -> int:
    await ctx.debug(f"[add] {a}+{b}")
    result = a+b
    await ctx.debug(f"result={result}")
    return result


if __name__ == "__main__":
    asyncio.run(
        mcp.run_async(
            transport="streamable-http", 
            host="0.0.0.0", 
            port=os.getenv("PORT", 8080),
        )
    )

add.py:

from fastmcp import Client

import asyncio
import google.oauth2.id_token
import google.auth.transport.requests
import os
import sys

args = sys.argv
if len(args) != 3:
    sys.stderr.write(f"Usage: python {args[0]} <a> <b>\n")
    sys.exit(1)

a = args[1]
b = args[2]

audience = os.getenv("CLOUD_RUN_URL")

request = google.auth.transport.requests.Request()
token = google.oauth2.id_token.fetch_id_token(request, audience)

config = {
    "mcpServers": {
        "cloud-run":{
            "transport": "http",
            "url": f"{audience}/mcp/",
            "headers": {
                "Authorization": "Bearer token",
            },
            "auth": token,
        }
    }
}

client = Client(config)


async def run():
    async with client:
        print("Connected")
        result = await client.call_tool(
            name="add",
            arguments={"a":a, "b":b},
        )
        print(result)


if __name__ == "__main__":
    asyncio.run(run())

Dockerfile:

# FastMCP Application Dockerfile
FROM docker.io/python:3.13-slim

RUN apt-get update && \
    apt-get install -y --no-install-recommends curl ca-certificates

ADD https://astral.sh/uv/install.sh /uv-installer.sh

RUN sh /uv-installer.sh && \
    rm /uv-installer.sh

ENV PATH="/root/.local/bin:${PATH}"

WORKDIR /app

COPY main.py main.py
COPY pyproject.toml pyproject.toml
COPY uv.lock uv.lock

RUN uv sync --locked

EXPOSE 8080

ENTRYPOINT ["uv", "run","/app/main.py"]

And:

BILLING="..."
PROJECT="..."

NAME="fastmcp"

REGION="..."

ACCOUNT="tester"
EMAIL=${ACCOUNT}@${PROJECT}.iam.gserviceaccount.com

gcloud iam service-accounts create ${ACCOUNT} \
--project=${PROJECT}

gcloud iam service-accounts keys create ${PWD}/${ACCOUNT}.json \
 --iam-account=${EMAIL} \
 --project=${PROJECT}

gcloud projects add-iam-policy-binding ${PROJECT} \
--member=serviceAccount:${EMAIL} \
--role=roles/run.invoker

gcloud auth print-access-token \
| podman login ${REGION}-docker.pkg.dev \
  --username=oauth2accesstoken \
  --password-stdin

REPO="cloud-run-source-deploy"
VERS="0.0.1"
IMAGE=${REGION}-docker.pkg.dev/${PROJECT}/${REPO}/${NAME}:${VERS}

podman build \
--tag=${IMAGE} \
--file=${PWD}/Dockerfile \
${PWD}

podman push ${IMAGE}

gcloud run deploy ${NAME} \
--image=${IMAGE} \
--region=${REGION} \
--project=${PROJECT} \
--no-allow-unauthenticated