mcp-knowledge-graph

mcp-knowledge-graph

4.3

mcp-knowledge-graph is hosted online, so all tools can be tested directly either in theInspector tabor in theOnline Client.

If you are the rightful owner of mcp-knowledge-graph and would like to certify it and/or have it hosted online, please leave a comment on the right or send an email to henry@mcphub.com.

Knowledge Graph Memory Server for AI models using Model Context Protocol.

The `mcp-knowledge-graph` is an advanced implementation of persistent memory using a local knowledge graph, allowing AI models to remember user information across interactions. It is designed to work with any AI model that supports the Model Context Protocol (MCP) or function calling capabilities. This server is a fork of the original Memory Server, offering a more robust solution by avoiding the ephemeral memory npx installation method. The server uses entities, relations, and observations to structure and store information, enabling AI models to maintain context and improve user interaction over time. The server is compatible with various AI platforms, including Claude, GPT, and Llama, and can be customized with a specific memory path for data storage.

Features

  • Persistent memory using a local knowledge graph
  • Customizable memory path for data storage
  • Compatible with any AI model supporting MCP or function calling
  • Entities, relations, and observations for structured information storage
  • Fork of the original Memory Server with improved implementation

Tools

  1. create_entities

    Create multiple new entities in the knowledge graph

  2. create_relations

    Create multiple new relations between entities in the knowledge graph. Relations should be in active voice

  3. add_observations

    Add new observations to existing entities in the knowledge graph

  4. delete_entities

    Delete multiple entities and their associated relations from the knowledge graph

  5. delete_observations

    Delete specific observations from entities in the knowledge graph

  6. delete_relations

    Delete multiple relations from the knowledge graph

  7. read_graph

    Read the entire knowledge graph

  8. search_nodes

    Search for nodes in the knowledge graph based on a query

  9. open_nodes

    Open specific nodes in the knowledge graph by their names