memories-with-lessons-mcp-server
43
The Knowledge Graph Memory Server is a tool for implementing persistent memory using a knowledge graph. It helps in storing and retrieving user information across sessions, and capturing and learning from errors.
Knowledge Graph Memory Server
A basic implementation of persistent memory using a local knowledge graph, allowing the Claude model to maintain user information across chats and learn from past errors. Key features include the management of entities, relations, observations, and lessons. The server offers tools such as creating entities, managing relations, adding observations, and more. It supports integration with Cursor MCP client and Claude Desktop with configurations for Docker and NPX.
Core Concepts
- Entities: Primary nodes with a unique name, type, and observations.
- Relations: Directed connections between entities.
- Observations: Pieces of information about entities.
- Lessons: Special entities to capture errors and solutions.
API Features
- Create and manage entities, relations, and observations.
- Lesson management with tracking success rates and recommendations.
- Integrations include building with Docker.