memento-mcp
Memento MCP is a robust knowledge graph memory system for language models, offering features like semantic retrieval, temporal awareness, and adaptive long-term memory. It uses Neo4j for storage and integrates seamlessly with various LLM clients, enhancing memory capabilities with advanced search and metadata features.
What is the primary storage backend for Memento MCP?
Memento MCP uses Neo4j as its storage backend, which provides both graph storage and vector search capabilities.
How does Memento MCP handle semantic search?
Memento MCP uses vector embeddings to perform semantic search, allowing it to find related entities based on meaning rather than just keywords.
Can Memento MCP track changes over time?
Yes, Memento MCP maintains a complete version history of entities and relations, enabling point-in-time graph retrieval and temporal consistency.
What is confidence decay in Memento MCP?
Confidence decay is a feature where the confidence in relations automatically decreases over time if not reinforced, based on a configurable half-life.
Is Memento MCP compatible with multiple LLM clients?
Yes, Memento MCP is compatible with any LLM client that supports the model context protocol, such as Claude Desktop, Cursor, and GitHub Copilot.