MCP Directory

Cloud Run MCP Server

Deploy apps and folders to Google Cloud Run directly from MCP-compatible AI agents.

Unverified
stdio (local)
OAuth
JavaScript
View repo 617 Website npm

Add to your client

Copy the config for your MCP client and paste it into its config file.

Install / run
npx -y @google-cloud/cloud-run-mcp

Paste into ~/Library/Application Support/Claude/claude_desktop_config.json

{
  "mcpServers": {
    "cloud-run-mcp-server": {
      "command": "npx",
      "args": [
        "-y",
        "@google-cloud/cloud-run-mcp"
      ]
    }
  }
}

Step-by-step guides: Add to Claude Desktop · Add to Cursor · Add to Windsurf

Before you start

  • Node.js (LTS recommended) for the Node.js setup, or Docker for the container setup
  • Google Cloud SDK (gcloud) installed and authenticated
  • Authenticate with `gcloud auth login` and `gcloud auth application-default login`
  • A Google Cloud account with access to Cloud Run / a GCP project

About Cloud Run MCP Server

An MCP server from Google Cloud that enables AI agents to deploy code to Cloud Run, list and inspect services, read logs, and manage GCP projects. It supports a local stdio setup with local gcloud credentials, a remote setup running on Cloud Run secured with IAM, and OAuth-based authentication. It can also be installed as a Gemini CLI extension.

Tools & capabilities (7)

deploy-file-contents

Deploys files to Cloud Run by providing their contents directly.

list-services

Lists Cloud Run services in a given project and region.

get-service

Gets details for a specific Cloud Run service.

get-service-log

Gets logs and error messages for a specific Cloud Run service.

deploy-local-folder

Deploys a local folder to a Google Cloud Run service (only available when running locally).

list-projects

Lists available GCP projects (only available when running locally).

create-project

Creates a new GCP project and attaches it to the first available billing account; a project ID can optionally be specified (only available when running locally).

What this server can do

Cloud Run MCP Server provides tools for these capabilities — tap one to see every MCP server that does the same:

Deploy & manage cloud infra

When to use it

  • Deploy the current working directory or local folder to Cloud Run from an AI-assisted IDE or assistant
  • Deploy app code to Cloud Run from Gemini CLI or other AI-powered CLI agents
  • List Cloud Run services and inspect a specific service's configuration
  • Fetch logs and error messages for a Cloud Run service to debug deployments
  • Create and manage GCP projects when running the server locally

Security notes

Local mode uses your Google Cloud application-default credentials, granting the server the deploy and project-management permissions of your gcloud identity. For remote mode, do NOT expose the server without authentication — the docs use IAM auth via `gcloud run services proxy` and deploy with `--no-allow-unauthenticated`. `SKIP_IAM_CHECK` defaults to `true` (makes deployed services public); set it to `false` to enforce IAM checks. `ENABLE_HOST_VALIDATION` (with `ALLOWED_HOSTS`) can be enabled to prevent DNS rebinding attacks. The tools `deploy-local-folder`, `list-projects`, and `create-project` are only available when running locally.

Cloud Run MCP Server FAQ

Is this for hosting MCP servers on Cloud Run?

No. This repository is an MCP server used to deploy your code to Cloud Run. To learn how to host MCP servers on Cloud Run, see the Cloud Run documentation referenced in the README.

How do I run it as a remote server safely?

Deploy the server to Cloud Run with `--no-allow-unauthenticated`, then run `gcloud run services proxy` on your machine to connect securely via your IAM identity. Point your MCP client at `http://localhost:3000/sse` (or use mcp-remote if your client lacks `url` support). Never expose the remote server without authentication.

Which tools require running locally?

`deploy-local-folder`, `list-projects`, and `create-project` are only available when the server runs locally.

Does it support OAuth?

Yes. The server supports OAuth as an authentication mechanism; configure a `.env` file (a `.env.example` and `.env.gcloud-sdk-oauth` are provided) and configure your MCP client for OAuth (e.g. Gemini CLI's `httpUrl` + `oauth` settings).

#google-cloud#cloud-run#deployment#devops#gcp#serverless

Alternatives to Cloud Run MCP Server

Compare all alternatives →
GitHub MCP Server
DevOps & Cloud
31k

GitHub's official server for repos, issues, PRs, and Actions — local Docker or hosted remote.

Featured
Verified
stdio (local)
API key
Go
7 tools
Updated 15 days agoRepo
AWS API MCP Server
DevOps & Cloud
9.3k

Official AWS Labs server that lets agents call any AWS API via the AWS CLI surface.

Verified
stdio (local)
API key
Python
3 tools
Updated 17 days agoRepo
kubefwd
DevOps & Cloud
4.1k

Bulk Kubernetes port forwarding with an MCP server that lets AI assistants forward services, inspect pods, and monitor traffic.

Unverified
stdio (local)
No auth
Go
28 tools
Updated 9 days agoRepo

Compare Cloud Run MCP Server with:

vs GitHub MCP Servervs AWS API MCP Servervs kubefwdvs Radar