Provides an embedded, local graph database for AI agents to store and retrieve contextual memories.
Sponsored
Neo Memory is a minimalist, local graph database solution designed to serve as persistent memory for AI coding agents. It utilizes an embedded KuzuDB for graph storage and local ONNX embeddings for efficient semantic recall, eliminating the need for cloud services, external servers, or complex setup. The tool offers a set of five core functionalities, including storing new memories, performing semantic searches, connecting related memories in a graph, deleting entries, and exploring memory relationships, thereby enhancing an agent's ability to retain context and learn across sessions.
主要功能
01Supports graph-based memory with explicit relationships between data points
