Knowledge Graph Memory Server

Example Use Cases

• 1

  Persistent, structured memory across conversations so they can store and recall user identity, preferences, and long-term facts.

• 2

  Model relationships between people, organizations, and events and query those connections during dialog.

• 3

  A lightweight, local SQLite-backed memory store that is easy to run via Docker or npx and integrate with Claude Desktop.

Description

This server implements a simple persistent memory system as a local knowledge graph (entities, relations, observations) so conversational agents can remember information across chats. It exposes an MCP-compatible API for creating/deleting entities and relations, adding/removing observations, searching nodes, and reading the whole graph. The project is designed for easy integration with Claude Desktop (Docker or npx) and is distributed under the MIT license so it can be used and modified freely. It is a fork/implementation intended for experimentation and demonstration rather than an official product.

Capabilities(9 total)