A message queue accepts messages from producers and delivers them to one or more consumers. The producer does not need the consumer to be available immediately, which helps systems absorb bursts of work and recover from temporary failures.
Agent runtimes use queues to schedule tasks, limit concurrency and retry failed processing. Correctness depends on acknowledgement, visibility timeout, duplicate handling and dead-letter behavior because delivery may occur more than once.
Acronyms and aliases
work queue synonymevent queue variant
Related terms
Frequently asked questions
Why do AI agent systems use message queues?
Queues decouple agents and tools, buffer uneven workloads and preserve tasks while a consumer is busy or temporarily unavailable.
Can a message queue deliver the same task twice?
Yes. Many queues provide at-least-once delivery, so consumers should be designed to handle duplicates safely.
