Hi Agent2Agent folks,

I've updated  draft-yao-agent-auth-considerations-01, Further
considerations on AI Agent Authentication and Authorization Based on OAuth
Extension.

https://datatracker.ietf.org/doc/draft-yao-agent-auth-considerations/

This draft primarily considers the AuthN and AuthZ issues for multiple
scenarios in agent communication networks, including agent to agent
communication, agent to API proxy server(MCP server) communication, and
delegated communication, etc.

-01 adds more security considerations, including  Agent Impersonation
during Discovery, Excessive Attribute Disclosure, and Token Replay and
Intermediary Exfiltration.

There might be more issues related and need to be considered.

Welcome for review and comments.

Best regards,
Kehan
_______________________________________________
OAuth mailing list -- [email protected]
To unsubscribe send an email to [email protected]

Reply via email to