I think 4 of the vulnerabilities, about skills, signature, identity are not willing to be treated by A2A voluntarily.

Because it need to be handled by the agent's owner themselves or a third party. Difficult to a protocol to guaranty that.

They are working on it with the AI Catalog and Trust Manifest.

This approach is interesting, probably near that SDD, but also expensive in time and tokens.

And it can lead, to my experience, to an other problem is looping between agents, that work without never finish on a scope by always identifying something, with often side effects that generate new problems.

Elon Musk said that (without him ;)) world demography gone a decline, as it is in Japan or Italy.

If radical right political parties arrive in Europe for exemple, are women gone a make children again? Or are we going to massively clone ourselves to conquer other planets?

Or finally let humanity slowing disappear letting robots replacing them definitively ?

What do you think?

Well. AI is built on data (so don't blame OpenAi).

We use to say in France: Americans innovate, the Chinese copy, and Europe regulates.

I'm curious, did you use open-source to build that?

Probably not teams from YC xD...or startup. I think scrum will die slowly.

There are many. For exemple LangGraph.

If I want to define a usefull workflow for myself, maybe I can only use skill?

Depends of usecase. What kind of actions do you want to perform with your agents ?

Maybe, I want to ask the LLM to work as a specific procegure, but the long Promt may lead the fault, if I can invite a "workflow" to confirm the LLM will work in proper manner that if they do somethings wrong the Harness can feedback and stop to execute the tool calls, there are much less mistakes.Perhaps...

Yes perhaps in this case you should watch the video of Garry Than about Personal AGI. He give a really interesting approach about creating your own context each days.

I personally start to apply it a few weeks ago.

Got it. I'll try it.

David Gurlé, founder of Antimatter, has an impressive track record of taking on industry giants with unconventional approaches in telecommunications.

He’s now taking a similarly contrarian approach to AI, betting on the decline of Anthropic and other proprietary AI players.

In a recent French-language interview with Carlos Diaz on Silicon Carne, he explains his vision for rethinking both AI infrastructure and energy consumption through mobile micro data centers.

I'am also dad of a seven month kid and in my case it increase dramatically my ambition.

I'm working harder again even if i have less time, because i want to build a legacy for him and want to show and make make him an exceptional life.

It can make you stronger.