Google Cloud Introduces Managed MCP Support: Enhance Your Cloud Experience

Introducing Google Cloud’s Managed Remote Model Context Protocol Servers

Google Cloud has just elevated its API infrastructure by rolling out fully managed Remote Model Context Protocol (MCP) servers. This innovative enhancement provides developers with a unified layer across Google and Google Cloud services, streamlining integration for various applications.

What Are MCP Servers?

With the launch of MCP servers, developers can now point their AI agents and standard MCP clients, such as the Gemini CLI, to a single, globally consistent, enterprise-ready endpoint. This integration simplifies access to Google’s suite of services, making it easier for developers to build and deploy applications across platforms seamlessly.

Incorporating MCP into Google’s Ecosystem

Google plans to incrementally roll out support for MCP across its services, which will initially include popular offerings like Google Maps, BigQuery, Google Compute Engine (GCE), and Google Kubernetes Engine (GKE). This phased approach allows developers to gradually implement the MCP capabilities without disruption and evaluate their effectiveness in real-world scenarios.

The Implications for Developers

The support for MCP servers is being hailed as a strong endorsement for this protocol, akin to USB-C for AI. However, it has also sparked discussions among developers. Some experts on platforms like Reddit have questioned whether cloud-based MCP solutions effectively address problems that local, trusted MCPs already solve, especially in terms of latency. There’s also ongoing debate about whether remote MCPs are merely transforming the protocol into a variant of remote APIs, similar to the traditional HTTP model.

Managing and Securing MCP Capabilities

To enhance management and security across services, Google has launched the Cloud API Registry and the Apigee API Hub. These tools are designed to help developers find and access trusted MCP tools from both Google and their organizations. Notably, Apigee can convert standard enterprise APIs—such as product catalogs—into discoverable MCP servers. This functionality allows companies to expose their custom business logic to AI agents while preserving vital governance and security frameworks.

Industry Collaboration and Support

Google’s initiative aligns with broader industry efforts such as the Agntcy project—a collaboration involving tech giants like Cisco, Oracle, Red Hat, and Dell Technologies. This project has been donated to the Linux Foundation, highlighting an industry consensus on the future of MCP. Additionally, prominent players like Amazon Web Services (AWS) and Microsoft are deeply engaged in the MCP ecosystem. AWS offers extensive resources through its Amazon Bedrock AgentCore, while Microsoft is incorporating MCP functionality directly into popular developer tools, including Visual Studio Code and Copilot.

Community Feedback and Developer Insights

In a recent post on Medium, Roman Irani remarked on the significance of Google’s managed MCP services. He noted that these advancements would significantly reduce friction points for developers working on agent development, particularly those integrating with various Google services. This streamlined interaction is expected to empower developers to innovate more rapidly and effectively.

Availability and Further Resources

Currently, MCP servers are in public preview, offering developers the opportunity to explore this new capability. A demo showcasing MCP integration with the supported Google services is available on GitHub, providing a practical resource for those eager to get started with these enhancements in their projects.