Google has launched its own fully managed remote MCP servers that will make it easier to integrate AI agents into Google and Cloud services like Maps and BigQuery. Instead of spending a week or two setting up connectors, developers can now simply paste a URL into a managed endpoint, the company says.
Google has launched its own fully managed remote MCP servers that will make it easier to integrate AI agents into Google and Cloud services like Maps and BigQuery. Instead of spending a week or two setting up connectors, developers can now simply paste a URL into a managed endpoint, the company says.
At launch, Google is offering MCP servers for Maps, BigQuery, Compute Engine, and Kubernetes Engine. In practice, this could look like an analytics assistant that makes direct queries to BigQuery, or an operations agent that interacts with infrastructure services, TechCrunch reports .
While MCP servers will eventually be available across all Google tools, they are currently in public preview, meaning they are not yet fully covered by the Google Cloud Terms of Service. However, they are available at no additional cost to enterprise customers who already pay for Google services.
“We expect to have general availability very soon, early in the new year,” said Steren Giannini, director of product management at Google Cloud. He added that new MCP servers will be released gradually, literally every week.
MCP (Model Context Protocol) was developed by Anthropic about a year ago as an open standard for connecting AI systems to data and tools. The protocol has gained widespread acceptance in the AI agent development community, and earlier this week Anthropic donated MCP to the new Linux Foundation , a foundation created to openly develop and standardize AI agent infrastructure.
MCP clients can be thought of as AI applications on the other end of the wire that talk to MCP servers and invoke the tools they offer. For Google, this includes Gemini CLI and AI Studio. Giannini noted that he has also tested Anthropic’s Claude and OpenAI’s ChatGPT as clients, and “they just work.”
Google says it's not just about connecting agents to its services. A bigger enterprise player is Apigee, their API management product, which many companies already use to issue API keys, set quotas, and monitor traffic.
Giannini said Apigee can essentially “convert” a standard API into an MCP server, turning endpoints like a product catalog API into tools that an agent can discover and use, with existing security and management controls layered on top of them.
In other words, the same API protections that companies use for human-built programs can now be applied to AI agents.
Google protects its MCP servers with an IAM access system that strictly restricts the actions of AI agents. An additional layer of security is provided by Google Cloud Model Armor, a special “shield” against AI-specific threats, such as attempts to hack model logic or steal data. In addition, all agent actions are recorded in audit logs, giving administrators full visibility into the process.
Google plans to expand MCP support beyond the initial set of servers. Over the next few months, the company will roll out support for services in areas such as data storage, databases, logging and monitoring, and security.
“We created a system so that developers don’t have to deal with it,” Giannini said.
