Let agents write a shared artifact concurrently; on conflict, notify affected agents and have them selectively repair, rather than locking or aborting.
Properties
family
Coordination & Scaling
maturity
frontier
diagram
shared-state
whyMultiAgent
Classical locks fit LLM agents poorly, a single “transaction” spans minutes of inference with broad, opaque read-sets, so concurrency control becomes notification-based rather than lock-based.