Persistent memory across sessions
Unlike standard chatbots that forget users the moment a session ends, Napster agents maintain memory across sessions and channels. When a returning user starts a new conversation, the agent already knows who they are and what was previously discussed.
How memory works
Each user interaction is associated with an identity. The platform stores conversation history and injects relevant past context at the start of new sessions — creating personalized experiences where the agent can reference prior conversations and build on previous progress.
Cross-channel consistency
Memory is consistent across channels. A user who previously spoke to the agent via video will be recognized if they return via voice or a new web session — the agent picks up exactly where it left off.
Tip: Session transcripts are accessible through the dashboard and Sessions API for review and auditing.