Engram is an open-source Model Context Protocol (MCP) server that provides AI agents with a universal memory layer. It does this by storing memories in local SQLite databases utilizing semantic vector search, knowledge graphs, and LLM-powered consolidation, enabling high-accuracy recall, spreading activation, and advanced memory management. It is designed for developers and teams building or orchestrating AI agents who need reliable, context-aware, and scalable memory beyond simple file-based approaches—without storing sensitive data in the cloud.
Visit Engram's official website for product details and getting started.
Comprehensive guides and API documentation for integrating and utilizing Engram.
Explore example implementations and projects using Engram for practical insights.