GCP MCP Server

Example Use Cases

• 1

  Query and inventory GCP resources (VMs, buckets, datasets) and return structured information for automation or reporting.

• 2

  Create, configure, or deploy services on GCP (e.g., deploy to Cloud Run, start/stop Compute Engine instances, trigger Cloud Build pipelines).

• 3

  Access logs, monitoring metrics, and audit information to diagnose issues or generate alert-driven remediation steps.

Description

This repository implements an MCP (Model Context Protocol) server exposing multiple Google Cloud services (Artifact Registry, BigQuery, Cloud Build, Compute Engine, Cloud Run, Cloud Storage, Monitoring, and Logging) to AI assistants. It centralizes client management, authentication (Application Default Credentials or service account), and service modules so agents can query infrastructure, create/configure resources, and receive AI-guided assistance. The server can be run locally, in development mode, or in Docker and is intended to provide a consistent interface for automating and managing GCP operations. Note: the README indicates it is not a production-ready MCP server and requires proper GCP project configuration and credentials.