MCP DuckDB Knowledge Graph Memory Server

Name: MCP DuckDB Knowledge Graph Memory Server
Author: MCP-Mirror

@MCP-Mirror

Repository

About MCP DuckDB Knowledge Graph Memory Server

Mirror of

#mathgpt #math-solver #math-assistant

Basic information

Config

Add this server to your MCP-compatible client using the configuration below.

{
  "mcpServers": {
    "IzumiSy_mcp-duckdb-memory-server": {
      "command": "npx",
      "args": [
        "-y",
        "@smithery/cli",
        "install",
        "@IzumiSy/mcp-duckdb-memory-server",
        "--client",
        "claude"
      ]
    }
  }
}

Tools

No tools detected

We auto-extract tools from the README. The maintainer can list them under a ## Tools heading to populate this section.

Overview

What is MCP DuckDB Knowledge Graph Memory Server?

MCP DuckDB Knowledge Graph Memory Server is a forked version of the official MCP Knowledge Graph Memory Server that replaces its JSON-based backend with DuckDB. It is designed for developers using MCP clients (e.g., Claude Desktop) who need persistent, scalable, and queryable memory storage for knowledge graphs.

How to use MCP DuckDB Knowledge Graph Memory Server?

Install automatically via Smithery (npx -y @smithery/cli install @IzumiSy/mcp-duckdb-memory-server --client claude) or manually by adding @Izumisy/mcp-duckdb-memory-server to claude_desktop_config.json with an optional MEMORY_FILE_PATH environment variable. A Docker image is also available. After setup, follow the provided interaction instructions to search and update memory during conversations.

Key features of MCP DuckDB Knowledge Graph Memory Server

Replaces JSON backend with DuckDB for better performance and scalability
SQL interface for complex, flexible queries
Transaction support for data integrity
Fuzzy search via Fuse.js on top of SQL results
Embedded database – no external server required
Supports indexing to improve search performance

Use cases of MCP DuckDB Knowledge Graph Memory Server

Persistent memory for AI assistants across chat sessions
Storing and querying large knowledge graphs with millions of entities
Enabling rich conditional searches (e.g., “find all goals related to user X”)
Maintaining atomic updates for concurrent MCP interactions

FAQ from MCP DuckDB Knowledge Graph Memory Server

How does this differ from the official Knowledge Graph Memory Server?

This version stores all data in a DuckDB database file instead of a JSON file, offering faster query performance, support for SQL queries, transaction safety, and better scalability for larger datasets.

What are the runtime requirements?

The server runs via npx (Node.js) and requires a Node.js runtime. Optionally, it can be used with Docker.

Where is the memory data stored?

Data is stored in a DuckDB database file. The path is set via the MEMORY_FILE_PATH environment variable (default location not specified).

Can it be used with other MCP clients besides Claude Desktop?

The README only provides configuration examples for Claude Desktop, but the server follows the MCP protocol and should work with any compatible MCP client.