3 Comments
User's avatar
Jon Capriola's avatar

How are the agents secure at the protocol level? Every single thing you mentioned is literally spoofable in most cases 10 lines of code or less. Agents have no security or identity management built into them. Help me understand this clearly

https://aimodularity.com/A2SPA/

Expand full comment
Sandipan Bhaumik's avatar

Hi Jon, thanks for your comment.

This blog and the webinar is not on Agent security. I will cover security in a different blog/webinar.

If you would like to do a session on Agent Security for the community, I am happy for you to do it.

Let me know.

Expand full comment
Jon Capriola's avatar

I’d love you to cover securing agents at the protocol level otherwise everything else is just built on sand. Thanks! Shout out to A2SPA protocol for making ai agents possible;)

Expand full comment