Cloud Run MCP

Cloud Run MCP is Google’s official open-source MCP server that lets AI agents deploy applications to Google Cloud Run. With it connected, an agent in your editor or terminal can ship code to a managed container service instead of only writing it. The project is released under the Apache-2.0 license, has about 633 stars on GitHub, and was actively updated in October 2026.

Overview

The server speaks the Model Context Protocol, so any MCP-compatible client can use it: the Gemini CLI, AI-powered IDEs, AI assistant apps and agent SDKs such as the Google Gen AI SDK and the Agent Development Kit. It exposes focused deployment tools: deploy file contents directly to a new or existing Cloud Run service, list the services in a project and region, fetch a service’s details and read its logs. Setup is a small JSON snippet that runs the server through npx with the @google-cloud/cloud-run-mcp package, and the project also ships as a Gemini CLI extension. Note the distinction the README makes: this server deploys apps to Cloud Run; hosting MCP servers themselves on Cloud Run is a separate documented topic.

Key Features

• Deploy applications to Cloud Run directly from MCP-compatible AI agents
• Works with Gemini CLI, AI-powered IDEs and AI assistant apps
• Tools to deploy file contents, list services, inspect services and read logs
• Also packaged as a Gemini CLI extension
• Simple npx-based setup, nothing extra to host
• Official Google Cloud Platform project, Apache-2.0 licensed

💳 Pricing

The Cloud Run MCP server is free and open source under the Apache-2.0 license. The applications you deploy run on Google Cloud Run, which is billed under Google Cloud’s normal pricing.

Who Is It For

Cloud Run MCP is for developers who already work with AI coding agents and want deployments to Google Cloud Run to happen from the same conversation as the coding.