Neo4j Agent Memory

A Neo4j-based memory system for AI agents that stores entities as nodes and relationships as edges in a graph database, enabling semantic memory management through word-tokenized search and temporal filtering. Built by KnowAll AI with contributions from Daichi Okazaki, it provides 10 specialized tools for creating memories (people, places, organizations), building connections between them, and querying the knowledge graph with configurable depth traversal. The implementation emphasizes an LLM-driven approach where the AI handles all intelligence while the server provides simple graph operations, automatic integer conversion, and comprehensive guidance tools, making it useful for conversational AI systems, voice assistants, and any AI agent requiring persistent contextual memory with relationship awareness.