MCP Server Memory

🚀 MCP Server: In-Memory Version

This project is a modified version of the original @modelcontextprotocol/server-memory package, tailored to specific needs. It implements the Model Context Protocol (MCP), providing AI models with persistent memory capabilities through a knowledge graph, enabling models like Claude to retain and recall information from interactions.

This project is derived from the original @modelcontextprotocol/server-memory package and has been modified according to personal requirements.

It is a server that implements the Model Context Protocol (MCP), providing persistent memory capabilities for AI models through a knowledge graph. It enables models (such as Claude) to retain and recall information from interactions.

✨ Features

This server implements the Model Context Protocol and serves as a bridge between AI models and the locally stored knowledge graph. It allows models to:

• Create and manage entities (people, places, concepts, etc.).
• Define relationships between entities.
• Store observations or facts related to entities.
• Search for and retrieve information from the knowledge graph.

By default, the knowledge graph is stored in the knowledge_graph.json file in the current working directory.

💻 Usage Examples

Compatibility with Claude Desktop

This server is designed to be compatible with MCP-compliant clients, such as the Claude Desktop application. You can configure it through the client settings.

Example mcpServers configuration:

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

...[Keep the rest of the existing content unchanged...]