googleapis/gcloud-mcp

What it solves

This project provides a set of Model Context Protocol (MCP) servers that allow AI assistants to interact with Google Cloud environments. It eliminates the need for users to memorize complex gcloud CLI command syntax, flags, and arguments by allowing them to manage cloud resources using natural language prompts.

How it works

The project implements the MCP specification to create a bridge between AI clients (like Claude Desktop, Cursor, or Gemini CLI) and Google Cloud services. It provides several specialized servers:

• gcloud: Executes gcloud CLI commands via natural language.
• observability: Queries logs, metrics, and traces using Google Cloud Observability APIs.
• storage: Manages Google Cloud Storage buckets and objects.
• backupdr: Manages Google Cloud Backup and Disaster Recovery resources.

Who it’s for

Cloud administrators, developers, and team members who want to simplify Google Cloud management and lower the barrier to entry for those less familiar with the gcloud CLI.

Highlights

• Natural Language Interface: Convert high-level descriptions of desired outcomes into cloud operations.
• Multi-Server Support: Dedicated servers for general CLI access, observability, and storage management.
• **Gemini Integration: First-class support for Gemini CLI and Gemini Code Assist.
• Permission Control: Permissions are tied to the active gcloud account, supporting service account impersonation for least-privilege access.
• Safety Restrictions: Prevents the execution of commands that would initiate interactive sessions or run arbitrary inputs unsuitable for AI agents.