mantyx

hey! Sorry - I just always wanted to write a piece with some interactivity so maybe I went too far with this one.

Here is a more static one: https://death-of-scrum.net/static/

I hope that helps!

There an aspect of this that I find it hard to describe in words but I will try here. The shift I observe is that we are moving from a step curve to a smoother curve; The development cycle has never been so fluid and focused on the outcomes than it is (Or should be) today.

Yes this is my observations. Task chunking, Sprints, MVPs - A lot of this exists because it makes the human process simpler to go through but now this is not the limitation anymore. Deciding what to do what will drive value was always the most important thing and it now became more critical than ever.

Having developed mantyx.io, I believe that rest apis are still champion. MCP is nothing more than rest wrappers most of the time and skills are cli wrappers which in turn are rest wrappers.

We are likely, at this point, preferring an implementation like this: https://www.ietf.org/archive/id/draft-cui-ai-agent-discovery... but also draft expiring in the next week or so.

Hey! I believe you commented on my other post with a similar thing. Right now I'm still observing the protocols being defined and trying to add compatibility layers and support as needed.

A2A don't seem to define a protocol for discovery yet - at least I couldn't find one. Happy to engage further and discuss if you want on what are you building.

One of our goals is to accelerate adoption of A2A protocols so anything that is happening on that domain is of our interest!

rafael@mantyx.io if you want to reach out and I agree that doing something like `m2a REGISTRY_URL` might be better generally or any other discovery mechanism.

That is 100% true. We are currently observing those specs and how they evolve to decide which and how to implement them. At the moment we support the ability to bring agents in house via a2a as well as build agents and expose them to third party via a2a.

The agent discoverability is next in line.

Great feedback as well so thanks for commenting!