MCP, A2A and the 12 Factor Agents

[Sorras]

Hello,

We have been seeing the emergence of some communication standards for agents, such as Anthropic's MCP (Model Context Protocol) and Google's (now Linux Foundation's) A2A (Agent to Agent).

I would like to explore how some principles of the 12-factor methodology correlate with these patterns.
For example, I see factor "06 - Launch, Pause, Resume" as closely related to the A2A stack, particularly regarding message specifications that initiate, pause, and resume a flow. Does that make sense?

Still drawing on factor 06, in my understanding...
The MCP aims to abstract and decouple the interaction between LLMs and external resources, especially tools, which—as described in the 12-Factor methodology—are essentially a way for non-deterministic code (LLMs) to execute deterministic code.
Based on this concept, would it be coherent to expose an agent through an MCP, considering the specific emergence of A2A for agent communication as its intended purpose?