Quantum Learning Machines When one agent hands a task to another, the receiving agent often needs to know who handed it over, on whose behalf, through which earlier agents, and which prior context the work depends on. It also needs to know that the handoff is fresh and meant for it, and the sender needs a clear answer back. This document describes a small, transport-neutral handoff context: a signed offer, a signed status reply, and the chain of earlier offers that led to the current one, defined in CDDL for JSON and CBOR, with a lifecycle that maps to Agent2Agent task states. Each offer names the one before it by hash, so that a dropped or reordered intermediate hop can be detected when the chain is carried. Prior context is named by digest only; any log or transparency service behind it is opaque. Appendices show how a SCITT receipt or a Learnlog signed head can be referenced. The document is offered as input to the agentproto working group's use cases and dialog-context work.