Interaction Dynamics Layer (IDL) 1.0: A Model-Agnostic Continuity Architecture for Long-Term AI Systems

Interaction Dynamics Layer (IDL) 1.0 introduces a model-agnostic continuity architecture for long-term AI systems based on three complementary forms of memory: communication memory (chat.log), semantic memory (index.log), and orientational memory (ev.log). The architecture standardizes the representation of interaction-induced orientational change through directed Evolution Vector (EV) objects while deliberately remaining independent of any specific extraction methodology or mathematical implementation. IDL is designed as a portable continuity layer that enables AI systems to preserve and continue long-term interactions without repeatedly processing complete conversation histories. The specification defines the continuity architecture itself rather than a particular implementation, allowing interoperability across models, platforms, and future analytical methods. This publication introduces the IDL architecture. The underlying Evolution Vector concept originates from the author's earlier Evolution Vector Memory (EVM) research, which is referenced separately.