memento-mcp

Overview

Memento MCP is a sophisticated knowledge graph memory system that provides LLMs with persistent, adaptive long-term memory. It uses Neo4j as its backend, enabling both graph storage and vector search capabilities. The system supports entities with rich metadata, relations with confidence scoring, and temporal tracking of all changes, allowing for point-in-time queries and historical analysis.

Try asking AI

After installing, here are 5 things you can ask your AI assistant:

• create_entities

  Create multiple new entities in the knowledge graph

  Try:Create entities for John Smith and Acme CorpTry:Add new entities to the knowledge graph
• add_observations

  Add new observations to existing entities

  Try:Add observations to John SmithTry:Update entity information with new details
• delete_entities

  Remove entities and their relations

  Try:Remove outdated entities from the knowledge graphTry:Delete entity records
• delete_observations

  Remove specific observations from entities

  Try:Remove incorrect observations from entitiesTry:Clean up entity information
• create_relations

  Create new relations between entities with enhanced properties

  Try:Create relations between entitiesTry:Connect entities in the knowledge graph
• get_relation

  Get a specific relation with its enhanced properties

  Try:Get relation between entitiesTry:Retrieve relationship information
• update_relation

  Update an existing relation with enhanced properties

  Try:Update relation strength between entitiesTry:Modify relationship information
• delete_relations

  Remove specific relations from the graph

  Try:Remove outdated relationsTry:Delete connections between entities
• read_graph

  Read the entire knowledge graph

  Try:Show me the entire knowledge graphTry:Display all entities and relations
• search_nodes

  Search for nodes based on query

  Try:Find specific entities in the graphTry:Search for nodes by name
• semantic_search

  Search for entities semantically using vector embeddings

  Try:Find semantically similar entitiesTry:Search related concepts
• get_entity_history

  Get complete version history of an entity

  Try:Show the history of an entityTry:View entity changes over time

Comparable tools

Installation

docker-compose up -d neo4j

{
  "mcpServers": {
    "memento": {
      "command": "npx",
      "args": ["-y", "@gannonh/memento-mcp"]
    }
  }
}

FAQ

What makes Memento different from other memory systems?

Memento uses a knowledge graph approach with Neo4j, providing temporal awareness, semantic search, and confidence decay features that traditional key-value memory systems don't offer.

Can I use it with other LLM clients besides Claude Desktop?

Yes, Memento MCP works with any LLM client that supports the Model Context Protocol, including Cursor and GitHub Copilot.

Compare memento-mcp with