The protocol proliferation problem: Making sense of the open agent stack
The Stack
etween MCP, A2A, and ARD, WTF: How should companies think about the protocols at the heart of agentic AI?
As usual, not everyone I talk to is on Bluesky (boo!), so some extra notes. We're used to having multiple protocols Papi Menon points out; Todd Segal and Junjie B. give me the technical details of how the three protocols complement each other and Matt Dickens explains how ARD not only improves MCP governance but also helps you avoid getting locked in to one model. It's David Mooter who gave me a mini maturity model for when to adopt each protocol (when you need it or you'll lose business by not having it).
The protocol proliferation problem: Making sense of the open agent stack
Between MCP, A2A, and ARD, WTF: How should companies think about the protocols at the heart of agentic AI?
- open agent protocols
- MCP
- A2A
- ARD
- ACP
- AIAF
- open governance
Did you enjoy this article?
Recommend it — Standard Reader surfaces well-loved writing to more readers across the network.