Google's open-source MCP server for enterprise databases — one binary that fronts AlloyDB, BigQuery, Spanner, Postgres, MySQL, Oracle, MongoDB, Snowflake and more.
Databases & DataCloud & Infrastructure
3 views
gcloud MCP Server
by Google
v0.5.3
Apache-2.0
Featured Cloud & Infrastructure
google-cloudgcloudclouddevopscliinfrastructure
Summary
Google's official MCP server for driving the gcloud CLI from natural language, with allowlist and denylist controls over which commands an agent may run.
Features
- Runs gcloud CLI commands from natural-language requests
- Chains multi-step Google Cloud operations into a single instruction
- Allowlist and denylist controls over which gcloud commands an agent may execute
- Installs as a Gemini CLI extension, or as a stdio server in any MCP client
- Sibling servers for Cloud Observability, Cloud Storage and Backup and DR
Installation
Set up this MCP server in your favorite AI agent — copy a ready-made configuration below.
Description
The gcloud CLI is enormous, and most of the friction in using it is remembering which of several hundred command groups, flags and argument shapes applies to the thing you want. Google's official gcloud MCP server removes that step: an agent describes the outcome, the server executes the corresponding gcloud command against your Google Cloud environment, and multi-step workflows can be chained into a single request instead of being assembled by hand.
The practical audience is wider than "people who dislike typing". Teams routinely have members who are competent engineers but not fluent in gcloud, and this server lets them perform real cloud operations without first learning the surface area. Because the server is a wrapper over the CLI rather than a bespoke API client, whatever your gcloud installation can do, the agent can do — which is also why the safety controls matter.
Command allowlists and denylists
The server supports allowlisting and denylisting gcloud commands, so an operations team can expose read-oriented commands to an agent while blocking destructive ones. For an MCP server that shells out to a cloud CLI, this is the single most important configuration to get right before pointing an agent at a production project.
Setup
For Gemini CLI and Gemini Code Assist, npx @google-cloud/gcloud-mcp init --agent=gemini-cli installs it as a Gemini CLI extension for the current user across all projects; gemini mcp list then shows it connected over stdio. Any other MCP client takes the standard stdio entry running npx -y @google-cloud/gcloud-mcp. Node.js 20 or higher and an installed, authenticated gcloud CLI are prerequisites.
Related servers
The same repository publishes sibling servers for Google Cloud Observability, Cloud Storage, and Backup and Disaster Recovery, and Google separately operates a set of remote, fully managed MCP servers for Google Cloud products that authenticate with your Google credentials over HTTP.
Apache 2.0 licensed and maintained by Google in the googleapis organisation.
Related MCP Servers
Datadog's official managed MCP server — query logs, metrics, traces, incidents and dashboards from an AI agent, inside your existing Datadog permissions.
Cloud & Infrastructure
Official Grafana MCP server: search and edit dashboards, query Prometheus, Loki and Pyroscope, and work with alerts, incidents and on-call from an agent.
Cloud & Infrastructure
1 views
Pulumi's official MCP server: search your Pulumi Cloud resources, read the provider registry while generating infrastructure code, and delegate work to the Pulumi Neo agent.
Cloud & Infrastructure