2026 Oct
This specification defines a standard framework for establishing, managing, and verifying the identity of software agents that operate within federated legal and commercial workflows. It addresses a fundamental question in the era of artificial intelligence: how does a machine participant prove who it is, what it is authorized to do, and who is accountable for its actions?
The standard is motivated by the operational reality that software agents are increasingly performing tasks that have legal and commercial consequences — executing transactions, accessing protected resources, presenting authorizations, and making decisions on behalf of human principals. These activities require a trust infrastructure that is at least as rigorous as the one applied to human participants but adapted for the distinct characteristics of machine actors: they operate continuously, can be instantiated and retired programmatically, may act across organizational boundaries, and must be governed at scale.
This specification provides that trust infrastructure through three interlocking capabilities:
Standards published by Mplify Alliance (formerly MEF Forum) are intended for general distribution to the public and may be downloaded from this site and reproduced without charge. Any reproduction of Mplify documents shall contain the following statement: “Reproduced with permission of Mplify Alliance.” All rights granted to Mplify under applicable copyright laws are expressly reserved. No permission is granted to any recipient or user of Mplify publications to modify any of the information contained therein and Mplify disclaims all responsibility and liability for such modifications.