Knowledge Graph Memory Server

What is Knowledge Graph Memory Server?

A persistent memory implementation using a local knowledge graph, allowing Claude to remember user information across chats and learn from past errors through a lesson system.

How to use Knowledge Graph Memory Server?

Install via npx or Docker, configure the MCP client (e.g., Claude Desktop or Cursor), and add a system prompt for memory usage. The server exposes tools for managing entities, relations, observations, and lessons. Environment variable MEMORY_FILE_PATH can set a custom storage path.

Key features of Knowledge Graph Memory Server

• Entities as primary nodes with name, type, and observations.
• Directed relations stored in active voice.
• Atomic observations attached to entities.
• Lesson system for error pattern capture and success tracking.
• Automatic file splitting when lines exceed 1000.

Use cases of Knowledge Graph Memory Server

• Persistent memory for personal assistant across chat sessions.
• Learning from past errors by storing lessons with solutions.
• Building a knowledge graph of user preferences, relationships, and goals.

FAQ from Knowledge Graph Memory Server

What data does the server store?

It stores entities (with type and observations), relations between entities, and lessons (error patterns, solutions, metadata) in local JSON files (memory.json and lesson.json).

How are lessons different from regular entities?

Lessons are special entities with structured error pattern information, solution steps, success rate tracking, and environmental context, designed for learning from errors.

What are the runtime requirements?

Requires Node.js and npm/npx or Docker. The server uses local file storage; no external database is needed.

How does file management work?

Files are automatically split if they exceed 1000 lines to maintain performance.

What transport/auth does the server use?

The server uses stdio transport (for Claude Desktop) and is intended for local use with no authentication mechanism described.