Knowledge Graph Memory Server
by MCP-Mirror
A basic implementation of persistent memory using a local knowledge graph. It allows Claude to remember user information across chats and learn from past errors through a lesson system.
Last updated: N/A
What is Knowledge Graph Memory Server?
This server is a knowledge graph-based persistent memory system designed to enhance the capabilities of language models like Claude. It stores entities, relations, and observations to maintain context across interactions and incorporates a lesson system for learning from past errors.
How to use Knowledge Graph Memory Server?
The server provides an API with tools for creating, reading, updating, and deleting entities, relations, and observations in the knowledge graph. It also includes tools for managing lessons, such as creating new lessons, finding similar errors, updating success rates, and getting lesson recommendations. The server can be integrated with Cursor MCP client or Claude Desktop by configuring the appropriate settings.
Key features of Knowledge Graph Memory Server
Persistent knowledge graph storage
Entity and relation management
Observation tracking
Lesson system for error learning
API for data manipulation
File management for memory and lesson data
Integration with Cursor MCP client and Claude Desktop
Use cases of Knowledge Graph Memory Server
Chat personalization
Error resolution and prevention
Contextual understanding
Long-term memory for AI agents
FAQ from Knowledge Graph Memory Server
What is an entity in the knowledge graph?
What is an entity in the knowledge graph?
An entity is a primary node in the knowledge graph, representing a person, organization, event, or any other concept. Each entity has a unique name, entity type, and a list of observations.
How are relations defined between entities?
How are relations defined between entities?
Relations define directed connections between entities, stored in active voice to describe how entities interact or relate to each other.
What are lessons and how are they used?
What are lessons and how are they used?
Lessons are special entities that capture knowledge about errors and their solutions, including error patterns, solution steps, success rate tracking, and environmental context.
How can I integrate this server with Claude Desktop?
How can I integrate this server with Claude Desktop?
You can integrate the server with Claude Desktop by adding the appropriate configuration to your claude_desktop_config.json file, specifying the command and arguments for running the server (either using Docker or NPX).
How does the server handle file management?
How does the server handle file management?
The server stores basic entities and relations in memory.json and lesson entities in lesson.json. Files are automatically split if they exceed 1000 lines to maintain performance.