Google Cloud MCP Server
@lreimer
Google Cloud MCP Server について
A MCP server implementation for Google Cloud using Go and Cobra.
基本情報
設定
以下の設定を使って、このサーバーを MCP 対応クライアントに追加してください。
{
"mcpServers": {
"gcp-mcp-server": {
"command": "npx",
"args": [
"@modelcontextprotocol/inspector",
"go",
"run",
"main.go"
]
}
}
}ツール
ツールは検出されませんでした
ツールは README から自動的に抽出されます。メンテナーは ## Tools という見出しの下に記載することで、このタブに反映できます。
概要
What is Google Cloud MCP Server?
A Model Context Protocol (MCP) server implementation for Google Cloud, built with Go and Cobra. It supports stdio and SSE transports and exposes operations for Projects (list, describe), Container Clusters (list, describe), and Cloud Run Services (list, describe).
How to use Google Cloud MCP Server?
Configure it as an MCP server in your client (e.g., Claude Desktop) by pointing to the binary with the --transport stdio argument. For local development, use the MCP inspector with either go run main.go (stdio) or go run main.go --transport sse (SSE). Deployment to Google Cloud Run is manual, using a Docker image or direct source build.
Key features of Google Cloud MCP Server
- Supports stdio and SSE transports
- Lists and describes Google Cloud projects
- Lists and describes Google Kubernetes Engine clusters
- Lists and describes Cloud Run services
- Built with Go and Cobra for easy extensibility
- Can be deployed to Google Cloud Run
Use cases of Google Cloud MCP Server
- Integrate Google Cloud resource queries into AI‑assisted workflows via MCP
- Allow an MCP client to inspect project, cluster, and Cloud Run details without manual gcloud commands
- Deploy as a remote SSE endpoint on Cloud Run for cloud‑native MCP access
FAQ from Google Cloud MCP Server
What dependencies or runtime does it require?
It is a Go binary with no external runtime beyond the operating system. For SSE mode, a running HTTP server is required; the README shows deployment to Google Cloud Run.
What transports and authentication are supported?
The server supports stdio and SSE transports. Authentication is handled via Google Cloud service account credentials, set as an environment variable (GOOGLE_APPLICATION_CREDENTIALS) or mounted as a secret on Cloud Run.
What data does it access?
It queries Google Cloud APIs for projects, GKE clusters, and Cloud Run services. No local data is stored; all responses come from live Google Cloud state.
Are there any known limits or missing operations?
Only three service areas are implemented: Projects (list, describe), Container (list, describe), and Cloud Run (list, describe). Other Google Cloud services are not included.
Can I run it locally without deploying to Cloud Run?
Yes. Use go run main.go locally with --transport stdio for stdin/stdout mode, or --transport sse to start an HTTP server on port 8000.
「クラウドとインフラ」の他のコンテンツ
Cloudflare MCP Server
cloudflareSupabase MCP Server
supabase-communityConnect Supabase to your AI assistants
AWS Model Context Protocol (MCP) Server
alexei-ledA lightweight service that enables AI assistants to execute AWS CLI commands (in safe containerized environment) through the Model Context Protocol (MCP). Bridges Claude, Cursor, and other MCP-aware AI tools with AWS CLI for enhanced cloud infrastructure management.
AWS MCP Servers
awslabsOpen source MCP Servers for AWS
🐳 docker-mcp
QuantGeekDevA docker MCP Server (modelcontextprotocol)
コメント