Shared Memory (Agent Context)
Also known as: Agent context, Shared state
A common state store that multiple AI agents read from and write to.
What is Shared Memory (Agent Context)?
Shared memory in multi-agent systems is the common state store agents use to coordinate. Without shared memory, each agent operates with only its own conversation history; with shared memory, agents from different vendors can pass partial results, context, and decisions. LeafMesh provides shared memory across all participating agents through its runtime.
Related concepts
Agent CoordinationMulti-Agent System
Looking for the agent operations fabric?
LeafMesh ADK is the production runtime for the concepts defined here.