Give your AI agents human-like memory
From the team behind Gatsby, Mastra is a framework for building AI-powered apps and agents with workflows, memory, streaming, evals, tracing, and Studio, an interactive UI for dev and testing. Start building: npm create mastra@latest
Memory integrates with Mastra's observability platform, allowing developers to trace which memories were recalled during agent execution, inspect memory retrieval patterns, and evaluate whether agents are accessing appropriate context for their decisions.
Observational Memory persists application state and conversation context, enabling agents to build understanding of user needs, preferences, and domain-specific information over time. This supports building internal copilots and customer-facing assistants that improve with interaction history.
Observational Memory stores conversation history and application state persistently, allowing agents to recall previous interactions and maintain context when users return. This enables coherent multi-session conversations where agents remember user preferences, past decisions, and conversation threads.
Memory Gateway manages token usage with configurable retrieval and storage, allowing agents to selectively recall relevant past context instead of including full conversation history in every prompt. This reduces redundant token consumption while maintaining conversation coherence.