s2005/mcp-memory-server
If you are the rightful owner of mcp-memory-server and would like to certify it and/or have it hosted online, please leave a comment on the right or send an email to henry@mcphub.com.
A basic implementation of persistent memory using a local knowledge graph.
The Knowledge Graph Memory Server is a standalone implementation extracted from the modelcontextprotocol/servers project. It provides a persistent memory solution using a local knowledge graph, allowing the AI model, Claude, to remember user information across different chat sessions. The server organizes data into entities, relations, and observations, enabling structured storage and retrieval of information. Entities are the primary nodes, each with a unique identifier, type, and list of observations. Relations define directed connections between entities, while observations are discrete pieces of information attached to entities. The server offers a comprehensive API for creating, managing, and querying the knowledge graph, supporting operations like creating entities and relations, adding observations, and searching nodes. It is designed to integrate seamlessly with platforms like Claude Desktop, offering flexible deployment options via Docker or NPX.
Features
- Persistent memory using a local knowledge graph
- Structured data storage with entities, relations, and observations
- Comprehensive API for managing and querying the knowledge graph
- Seamless integration with platforms like Claude Desktop
- Flexible deployment options via Docker or NPX
