google/mcp

Google 💚 MCP

What it solves

This repository provides a centralized hub for Google's official Model Context Protocol (MCP) servers, enabling AI agents to seamlessly connect to and interact with Google's ecosystem of data sources and cloud infrastructure. It solves the problem of integrating AI models with real-world tools and live data from services like BigQuery, Google Maps, and Google Workspace.

How it works

The project organizes MCP servers into two main categories:

  • Remote MCP Servers: Managed by Google and available via endpoints, allowing for immediate integration with services like AlloyDB, Bigtable, and Cloud Storage.
  • Open-source MCP Servers: Servers that can be run locally or deployed to Google Cloud (e.g., via Cloud Run or GKE), covering tools like Firebase, Google Analytics, and Chrome DevTools.

It also includes examples (such as "Launch My Bakery" and "Allstrides") and guidance on how to host and secure these servers on Google Cloud infrastructure.

Who it’s for

Developers building AI agents and LLM-powered applications that need to integrate with Google Cloud services, Google Workspace, or other Google developer tools.

Highlights

  • Broad Ecosystem Integration: Connects AI agents to a wide array of Google services including databases (BigQuery, Spanner, Cloud SQL), infrastructure (GKE, Compute Engine), and productivity tools (Gmail, Docs, Sheets).

  • Cloud-Native Deployment: Detailed guidance and codelabs for deploying MCP servers to Cloud Run and GKE for production-ready hosting.

  • Ready-to-Use Examples: Sample agents built with the Agent Development Kit (ADK) demonstrating real-world use cases.

  • Diverse Tooling: Includes servers for specialized tasks like Google Maps Grounding Lite, Google Cloud Security, and Flutter/Dart development.

Related

  • Project
    taylorwilsdon/google_workspace_mcp

    A comprehensive Model Context Protocol (MCP) server that gives AI assistants full natural language control over 12 Google Workspace services, including Gmail, Drive, and Calendar.

  • Project
    GoogleCloudPlatform/cloud-run-mcp

    An MCP server that enables AI agents to deploy and manage applications on Google Cloud Run through a standardized protocol.

  • Project
    microsoft/mcp

    A collection of Model Context Protocol (MCP) servers and core tooling that standardizes how AI agents and LLMs connect to Microsoft services and external data sources.

  • Project
    IBM/mcp

    A collection of Model Context Protocol (MCP) servers and tools that enable AI agents to securely interact with IBM products and enterprise resources.

  • Project
    github/github-mcp-server

    A Model Context Protocol (MCP) server that connects AI agents and assistants to GitHub, enabling them to read code, manage issues, and automate workflows via natural language.