Skip to main content
Athena agents are conceptualized with memories, a feedback system that collects and remembers historical conversations to optimize their responses and behavior. This systematic approach to gathering and using conversation history allows the agent to continuously evolve, making real-time adjustments to better serve your needs while maintaining a historical record of improvements and trends.
Agent memory and saved session history are separate. Memory settings control collection and recall across conversations; disabling memory does not mean the session transcript is not stored.
See Agents for configuration and tools.
Athena leverages an advanced knowledge graph structure to organize and interconnect memories, enabling more intelligent retrieval and contextual understanding.
Knowledge graph showing linked memory nodes