The Agentic AI Foundation (AAIF), a neutral organization supporting the development of the agentic AI technology stack, has introduced the Model Context Protocol Associate (MCPA). The new vendor-neutral certification is designed to align with emerging roles in AI engineering, platform engineering, and AI governance. The announcement comes as the global AI community gathers in Amsterdam this week for AGNTCon + MCPCon Europe.
Connecting AI models to the tools and services they need has become a critical part of building agentic systems, and MCP has become the standard for enabling those connections. MCP’s usage has grown at an astonishing rate. Across MCP’s Tier 1 SDKs, monthly downloads are approaching half a billion, with both the TypeScript and Python SDKs surpassing 1 billion downloads total. Usage in production is climbing just as fast. MCP tool calls from ChatGPT users reached 98 times their January level by August, and Resend passed one million MCP calls in a single month. That scale is exactly why a common way to validate MCP expertise is needed. The MCPA provides engineers, platform teams, and AI governance professionals a way to demonstrate their understanding of how the protocol works, how its components interact, and how to apply it responsibly.
“We built MCP so developers could rely on one open protocol instead of writing custom integrations for every system. Since then, it has grown exponentially. Giving developers the option to earn a certification in MCP will formalize a shared understanding of how the protocol works, while ensuring this expanding ecosystem continues to operate smoothly,”
-David Soria Parra, co-creator and lead maintainer of MCP
The MCPA is a 120-minute, online, proctored, multiple-choice exam covering five domains: MCP Fundamentals (16%), Architecture & Components (14%), Interactions & Execution (26%), Security & Governance (24%), and Use Cases & Ecosystem (20%). Candidates are tested on MCP hosts, clients, servers, and tools; message flow and protocol primitives; trust boundaries, permissions, and risk controls; and how MCP is applied across real-world agentic deployments. The certification is aligned with the latest MCP specification release, 2028-07-28, and is recommended for candidates who understand the internals of the MCP and are comfortable reasoning about how the protocol works and how its components communicate.
“MCP has quickly become the standard for connecting AI agents to external systems. As organizations increasingly adopt agentic AI, they need developers who understand how those connections work and how to implement them responsibly, including the permissions and trust boundaries involved. With the MCPA, developers can demonstrate that knowledge through an official certification, giving employers a shared benchmark for assessing the skills required for modern software development.”
-Angie Jones, Vice President of Developer Experience, Agentic AI Foundation
The MCPA is now available for enrollment. Attendees of AGNTCon + MCPCon Europe in Amsterdam this week can register at an exclusive 20% discount. The same 20% savings are also being offered to participants registering for AGNTCon + MCPCon North America, scheduled for October 22–23, 2026, in San Jose, California.
The Agentic AI Foundation (AAIF) is the neutral home where the agentic stack is being built. With founding projects including MCP, A2A, AGENTS.md, goose, agentgateway, and Agent Router, AAIF governs the core standards and protocols that enable agents to operate interoperably across platforms. Through transparent governance and broad industry participation, AAIF is driving adoption and ensuring agentic AI infrastructure evolves openly, predictably, and at production scale.
