Skip to main content
Glama
bmandalcusat

MCP Boilerplate

by bmandalcusat
OverviewSchemaSourceRelated ServersDiscussionsAdmin
Python
Hybrid

MCP Boilerplate

This repository contains the code to demonstrate MCP capabilities

Build Status Python 3.10+ FastAPI Code style: black

Setup instructions

–2. Create a virtual environment and setup dependencies

python3 -m venv .venv
source .venv/bin/activate 
pip install -r requirements.txt

Related MCP server: Cloudflare Google OAuth MCP Server

To deploy Remote MCP

We're going to deploy a remote MCP server to google cloud which also supports authentication. Navigate to the server remote-mcp-gcp directory and perform the following steps

  1. Install uv on your computer if you'd like to run the invoke the MCP server locally using your testing script.

  2. Rename .envrc_sample to .envrc so that your shell can pickup the GCLOUD_PROJECT_ID environment variable mv .envrc_sample .envrc

  3. Update your GCLOUD_PROJECT_ID in your .envrc file with the ID of your google cloud project.

  4. Create a container registry to host your MCP server container image

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

  1. Submit a build job for the container image. We'll use remote build for this.

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

  1. Create a container cloud run instance with

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

  1. Create a proxy to invoke the remote endpoint from your computer with authentication. gcloud run services proxy mcp-server --region=us-central1. This will ask you to install cloud run proxy tooling on your computer.

  2. Now, localhost:8080 should be pointed to your deployed instance with authentication enabled

  3. Run uv run test_server.py to invoke the client script against the remote server with various tools.

Feel free to create new tools and experiment.

IMPORTANT

Once you're done, delete all associated resources from Google Cloud to avoid unnecessary charges.

FAQ

  1. FastMCP vs MCP Python SDK.Read this issue for more info but generally FastMCP is much more preferred by developers and you'll be able to build much more capabilties with the same.

  2. Stop converting your REST APIs to MCP

This server cannot be deployed

One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Maintenance

ActivityInactive
ResponsivenessNo issues

Resources

Related MCP Connectors

Related MCP Servers

  • Cloud Run MCP Server

    vishal84
    F
    license
    Not graded
    quality
    C
    maintenance
    A secure MCP server example that demonstrates how to deploy to Google Cloud Run with authentication and identity token protection. Serves as a tutorial template for building production-ready MCP servers in the cloud.
    -

  • Cloudflare Google OAuth MCP

    zxzinn
    F
    license
    Not graded
    quality
    D
    maintenance
    A remote MCP server deployable on Cloudflare Workers that provides integrated Google OAuth authentication for secure tool access. It serves as a reference implementation for building persistent, authenticated MCP connections using Cloudflare Durable Objects and Streamable HTTP.
    1
    -

  • MCP Boilerplate

    anshajk
    A
    license
    Not graded
    quality
    D
    maintenance
    A boilerplate for deploying a remote MCP server on Google Cloud with authentication, enabling experimentation with MCP tools.
    8
    MIT