Agentic memory is an agent's ability to remember things. Short-term memory holds context within a single conversation or task — what was said, what was decided, where things stand. Long-term memory persists across sessions, storing user preferences, past interactions, and decisions made in previous runs.
Without memory, every new interaction starts from scratch regardless of history. A support agent that remembers a user reported an issue two days ago can pick up where things left off rather than asking them to explain the problem again. That continuity is the difference between an AI that feels useful and one that feels frustrating.
