OpenAI and Google Adopt Model Context Protocol to Unify AI Agent Deployments

The Model Context Protocol (MCP) is an open standard that links large‑language‑model applications to external tools and data. OpenAI incorporated MCP into its Agents SDK in March 2025, and Google added native MCP support to the Gemini API. The protocol, now overseen by the Agentic AI Foundation at the Linux Foundation, is used for developer tooling, knowledge‑worker copilots, and self‑serve analytics, though enterprises are still addressing governance and access‑control gaps.

Separately, organizations face fragmentation when deploying agents across multiple runtimes such as Claude Managed Agents, Cursor, AWS Bedrock and custom platforms. The proposed Control Plane pattern separates the governance layer from the runtime, providing a unified gateway that normalises invocation, centralises credentials, enforces uniform policies, maintains session continuity and enables cross‑team sharing of agents. Implementations like LiteLLM Agent Platform or Microsoft Foundry illustrate this approach.