Memento MCP
by gannonh
Memento MCP is a scalable knowledge graph memory system for LLMs, providing semantic retrieval, contextual recall, and temporal awareness. It offers LLM clients resilient, adaptive, and persistent long-term ontological memory.
Last updated: N/A
What is Memento MCP?
Memento MCP is a knowledge graph memory system designed to enhance LLMs with long-term memory capabilities. It uses Neo4j as its storage backend and provides features like semantic search, temporal awareness, and confidence decay to manage and retrieve information effectively.
How to use Memento MCP?
To use Memento MCP, you need to set up a Neo4j database (either locally with Neo4j Desktop or using Docker). Configure the necessary environment variables for database connection and OpenAI API key for semantic embeddings. Integrate Memento MCP with an LLM client that supports the Model Context Protocol, such as Claude Desktop, by configuring the claude_desktop_config.json file.
Key features of Memento MCP
Scalable knowledge graph storage using Neo4j
Semantic search with vector embeddings
Temporal awareness with version history
Confidence decay for relations
Advanced metadata support for entities and relations
MCP API tools for entity and relation management, graph operations, and semantic search
Use cases of Memento MCP
Providing LLMs with persistent memory of past conversations
Enabling LLMs to retrieve relevant information based on semantic meaning
Tracking the evolution of knowledge over time
Managing relationships between entities with varying degrees of confidence
FAQ from Memento MCP
What is the purpose of Memento MCP?
What is the purpose of Memento MCP?
Memento MCP provides LLMs with a long-term memory system based on a knowledge graph, enabling them to retain and retrieve information more effectively.
What is Neo4j and why is it used?
What is Neo4j and why is it used?
Neo4j is a graph database that provides a unified solution for graph storage and vector search, making it ideal for Memento MCP's knowledge graph implementation.
How do I set up Neo4j for Memento MCP?
How do I set up Neo4j for Memento MCP?
You can set up Neo4j either using Neo4j Desktop or with Docker. Instructions for both methods are provided in the README.
How do I configure Memento MCP with Claude Desktop?
How do I configure Memento MCP with Claude Desktop?
You need to add a configuration block to your claude_desktop_config.json file, specifying the command, arguments, and environment variables for Memento MCP.
What environment variables are required for Memento MCP?
What environment variables are required for Memento MCP?
The required environment variables include Neo4j connection settings (URI, username, password, database), OpenAI API key, and vector search configuration parameters.