What is the Google Cloud MCP Server?

It's a Go-based implementation of an MCP (Model Configuration Protocol) server for managing Google Cloud resources. It provides a standardized way to interact with GCP services.

What Google Cloud resources does it currently support?

It currently supports Projects, Container Clusters, and Cloud Run Services. You can list and describe these resources using the server.

How can I deploy the Google Cloud MCP Server?

Currently, manual Google Cloud Run deployment is used, either directly from source or using the Docker image built on Github. The README provides detailed steps.

What transport methods does the MCP server support?

The server supports both stdio and SSE (Server-Sent Events) as transport mechanisms for communication.

What is Cobra CLI used for?

Cobra is a framework used to build powerful modern CLI applications in Go. It is used to create the command-line interface for the MCP Server.

Google Cloud

Name: Google Cloud
Author: lreimer

bylreimer

•

Herramientas para Desarrolladores

Infraestructura en la Nube

Despliegue y DevOps

Implements an MCP server for Google Cloud resources using Go.

This tool provides an implementation of an MCP (Model Context Protocol) server tailored for Google Cloud. Built with Go and the Cobra framework, it facilitates the management and introspection of Google Cloud resources. It supports both `stdio` and `SSE` transport mechanisms, offering flexibility in how the server interacts with clients. Currently, it supports operations for listing and describing Projects, Container Clusters, and Cloud Run Services, streamlining the process of querying and understanding your Google Cloud infrastructure.

Características Principales

01Implements Container Clusters List and Describe operations.

021 GitHub stars

03Implements Cloud Run Services List and Describe operations.

04Supports stdio and SSE transport.

05Implements Projects List and Describe operations.

06Built with Go and Cobra.

Casos de Uso

01Local development using MCP introspector.

02Google Cloud Run deployment directly from source or Docker image.

03Local development with tools like Claude Desktop.