xgmem

meetdhanani17/xgmem

3.4

If you are the rightful owner of xgmem 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.

xgmem is a TypeScript-based Model Context Protocol (MCP) server designed to provide project-specific and knowledge graph-based memory for Claude, LLM agents, and other tools.

xgmem is a versatile MCP server that facilitates the storage, retrieval, and management of entities, relations, and observations on a per-project basis. It is built with TypeScript and supports Docker, making it a flexible solution for agents and LLMs requiring structured memory. The server is designed to enable cross-project knowledge sharing and migration, with a focus on scalability and persistence. xgmem stores memory in a JSON file, ensuring data is disk-persistent and queryable, which is crucial for maintaining a robust agent ecosystem.

Features

  • Knowledge graph storage for entities, relations, and observations
  • CRUD operations via MCP tools
  • Persistence to disk (memory.json)
  • Docker and TypeScript support

Usages

usage with mcp config

"mcpServers": {
    "xgmem": {
      "command": "npx",
      "args": ["-y", "xgmem@latest"]
    }
  }

usage with docker

sh
docker build -t xgmem-mcp-server .
docker run -v $(pwd)/memories:/app/memories xgmem-mcp-server

Tools

  1. save_project_observations

    Saves observations (memory) for a project.

  2. get_project_observations

    Retrieves observations for a specific project.

  3. add_graph_observations

    Adds observations to the knowledge graph.

  4. create_entities

    Creates new entities in the knowledge graph.

  5. create_relations

    Creates new relations between entities.

  6. delete_entities

    Deletes entities from the knowledge graph.

  7. delete_observations

    Deletes observations from a project.

  8. delete_relations

    Deletes relations between entities.

  9. read_graph

    Reads the entire knowledge graph.

  10. search_nodes

    Searches for nodes in the knowledge graph.

  11. search_all_projects

    Searches across all projects for specific data.

  12. open_nodes

    Opens nodes for detailed inspection.

  13. copy_memory

    Copies memory from one project to another.