mcp-server-memory

mcp-server-memory

3.4

An MCP server providing persistent memory capabilities for AI models through a knowledge graph.

The MCP Server: Memory is a specialized server that implements the Model Context Protocol to facilitate persistent memory capabilities for AI models. It acts as a bridge between AI models and a locally stored knowledge graph, allowing models to create and manage entities, define relationships, store observations, and retrieve information. This server is particularly useful for AI models like Claude, enabling them to retain and recall information across different interactions. The knowledge graph is stored in a JSON file, making it easy to manage and update. This server is derived from the original @modelcontextprotocol/server-memory package and has been modified to meet specific needs.

Features

  • Persistent Memory: Retains information across interactions using a knowledge graph.
  • Entity Management: Allows creation and management of entities such as people, places, and concepts.
  • Relationship Definition: Enables defining relationships between different entities.
  • Information Storage: Stores observations or facts associated with entities.
  • Information Retrieval: Provides search and retrieval capabilities from the knowledge graph.

Usage with Different Platforms

Claude Desktop


{
  "mcpServers": {
    "memory": {
      "command": "npx",
      "args": ["-y", "mcp-server-memory"],
      "env": {
        "MEMORY_FILE_PATH": "/path/to/your/custom_memory.json"
      }
    }
  }
}