Agent memory
Agent memory is the ability of an AI agent to store and recall facts, decisions, and outcomes across runs, so it improves over time instead of restarting from scratch each session.
In detail
Agent memory spans short-term working memory within a run and long-term memory that persists after it. Long-term agent memory requires external storage the agent writes to and queries. When agent memory is shared across a team rather than scoped to a single user, every agent run benefits from the whole team's accumulated context.