Hi OAuth WG,

We have published a new version of 
draft-li-oauth-delegated-authorization<https://datatracker.ietf.org/doc/html/draft-li-oauth-delegated-authorization-01>
 (GitHub : 
liuchunchi/li-oauth-delegated-authorization<https://github.com/liuchunchi/li-oauth-delegated-authorization>).

This draft defines a mechanism for clients (e.g. AI agents or user-controlled 
apps) to delegate a subset of their granted privileges to other parties (e.g. 
AI agents) in subordinate tokens, enabling fine-grained access control 
delegation while maintaining security and privacy.

Key updates since draft-00:


  *   Privacy Considerations - Added comprehensive analysis of privacy benefits 
including minimized authorization server visibility, no access pattern 
correlation, reduced data collection, and network traffic reduction. Also 
discusses trade-offs compared to Token Exchange (RFC 8693).
  *   Step-up Authorization Integration - Added Appendix B describing 
integration with step-up authorization challenge.

We are also planning a hackathon 
project<https://wiki.ietf.org/en/meeting/125/hackathon#oauth-delegated-authorization-for-ai-agents>
 at IETF 125 in Shenzhen that aims to implement this draft in AI agents.

Feedback and comments are welcome.

Best Regards,
Li Ruochen
_______________________________________________
OAuth mailing list -- [email protected]
To unsubscribe send an email to [email protected]

Reply via email to