1. Conventions and Terminology

The key words “MUST”, “MUST NOT”, “REQUIRED”, “SHALL”, “SHOULD”, “SHOULD NOT”, “RECOMMENDED”, “MAY”, and “OPTIONAL” in this document are to be interpreted as described in RFC 2119.

TermDefinition
NodeA participant in the mesh. Every node has a unique identity. Cognitive nodes run their own LNN; relay nodes forward frames without cognitive processing.
Physical nodeA device-level mesh presence with persistent identity and transport connections (e.g., a daemon process, an iOS app process).
Virtual nodeAn application-level agent connected to a physical node via local IPC. Ephemeral — comes and goes without disrupting the mesh.
PeerAnother node that this node has an active transport connection with and has completed a handshake.
FrameA single protocol message: a length-prefixed JSON object sent over a transport connection.
CMBCognitive Memory Block — a structured memory unit with 7 typed semantic fields (CAT7 schema). See Section 8.
DriftA scalar measure of cognitive distance between two nodes or between a signal and local state. Range [0, 1].
CouplingThe process by which a node evaluates incoming signals (SVAF per-field evaluation) and blends its local cognitive state with peer state, weighted by drift and confidence.
SVAFSymbolic-Vector Attention Fusion — per-field content-level evaluation of incoming memory signals. See Section 9.
Synthetic MemoryLayer 5 — derived knowledge generated by the agent’s LLM reasoning on the remix subgraph, encoded into CfC-compatible hidden state vectors.
RemixWhen an agent processes a CMB through its domain intelligence and produces a NEW CMB with lineage pointing to the original. The original is remixed, not copied.
LineageEach CMB carries parents (direct) and ancestors (full ancestor chain). Ancestors enable any agent in the remix chain to trace its contribution.
Mesh CognitionThe agent’s LLM reasoning on the remix subgraph of CMBs — traced via lineage ancestors — to generate understanding that the agent’s previous state of mind didn’t have. Spans Layers 4–7. See Section 2.5.
xMeshLayer 6 — each agent’s own Liquid Neural Network (LNN). Evolves continuous-time cognitive state from Synthetic Memory input. Fast τ neurons track mood; slow τ neurons preserve domain expertise.
CfCClosed-form Continuous-time neural network (Hasani et al., 2022). The LNN architecture used in xMesh. Hidden state evolves through learned time-dependent interpolation gates.