MCP DuckDB Memory
A memory server for Claude that stores and retrieves knowledge graph data in DuckDB, enhancing performance and query capabilities for conversations with persistent user information.
What is MCP DuckDB Memory?
MCP DuckDB Memory is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to memory server for claude that stores and retrieves knowledge graph data in duckdb, enhancing performance and query capabilities for conversations with persistent user information.
A memory server for Claude that stores and retrieves knowledge graph data in DuckDB, enhancing performance and query capabilities for conversations with persistent user information.
This server falls under the Knowledge & Memory and Databases categories on MCPgee, the world's largest MCP server directory with 33,000+ servers.
Features
- A memory server for Claude that stores and retrieves knowled
Use Cases
Maintainer
Works with
Installation
Manual Installation
npx mcp-duckdb-knowledge-graph-memory-serverConfiguration
Configuration Details
claude_desktop_config.json
Performance
Response Metrics
Resource Usage
How to Set Up and Use MCP DuckDB Memory
The MCP DuckDB Knowledge Graph Memory Server gives Claude a persistent, queryable memory by storing entities, observations, and relationships in a local DuckDB database. Unlike in-memory or flat-file memory solutions, it uses DuckDB's SQL engine for efficient storage and retrieval, plus Fuse.js-powered fuzzy search for finding entities even with partial name matches. Claude can remember facts about users, preferences, goals, and relationships across conversations — making interactions feel continuous rather than starting from scratch each session.
Prerequisites
- Node.js 18 or higher and npm/npx installed
- An MCP-compatible client such as Claude Desktop
- Write access to a local directory for the DuckDB database file (optional: set via MEMORY_FILE_PATH)
Install via Smithery (recommended)
The easiest way to install is through Smithery, which automatically configures Claude Desktop.
npx -y @smithery/cli install @IzumiSy/mcp-duckdb-memory-server --client claudeOr configure manually in claude_desktop_config.json
If you prefer manual setup, add the server directly to your Claude Desktop configuration file without a separate install step.
{
"mcpServers": {
"graph-memory": {
"command": "npx",
"args": ["-y", "@izumisy/mcp-duckdb-memory-server"]
}
}
}(Optional) Set a persistent database file path
By default, the server uses a temporary or default location. Set MEMORY_FILE_PATH to a stable location so your knowledge graph persists across server restarts.
{
"mcpServers": {
"graph-memory": {
"command": "npx",
"args": ["-y", "@izumisy/mcp-duckdb-memory-server"],
"env": {
"MEMORY_FILE_PATH": "/Users/yourname/.claude/memory.duckdb"
}
}
}
}Restart Claude Desktop
Restart the app after updating the config file. The server will initialize the DuckDB schema on first run.
Prime Claude with a memory system prompt
Add a system prompt to Claude instructing it to use the memory server. The README suggests prompting Claude to identify users by name, retrieve relevant memories at the start of each conversation, and update entities when new information is shared.
MCP DuckDB Memory Examples
Client configuration
Full configuration with a persistent database file path for claude_desktop_config.json
{
"mcpServers": {
"graph-memory": {
"command": "npx",
"args": ["-y", "@izumisy/mcp-duckdb-memory-server"],
"env": {
"MEMORY_FILE_PATH": "/Users/yourname/.claude/memory.duckdb"
}
}
}
}Prompts to try
Prompts that exercise the entity, observation, and relationship memory tools
- "Remember that my name is Alex and I prefer TypeScript over JavaScript"
- "What do you know about me from our previous conversations?"
- "Update my profile: I'm now working on a new project called Stellar at a fintech startup"
- "Add a note that I prefer morning stand-ups and avoid meetings after 3pm"
- "What are my current goals that you've recorded?"
- "Search your memory for anything related to my Python projects"Troubleshooting MCP DuckDB Memory
Memory is lost between Claude Desktop sessions
Set the MEMORY_FILE_PATH environment variable to a fixed path on your filesystem (e.g., ~/.claude/memory.duckdb). Without this, the server may use a temporary location that is cleared between runs.
npx fails with 'Cannot find package @izumisy/mcp-duckdb-memory-server'
The package name uses lowercase 'izumisy'. Ensure the args array in your config uses '@izumisy/mcp-duckdb-memory-server' (all lowercase). Run 'npx -y @izumisy/mcp-duckdb-memory-server' in your terminal to pre-cache it.
Fuzzy search returns irrelevant entities
Fuse.js fuzzy search is threshold-based. If results are too loose, try being more specific in your queries. If results are too strict and miss obvious matches, this is a known trade-off of the fuzzy algorithm — try exact entity names when precision matters.
Frequently Asked Questions about MCP DuckDB Memory
What is MCP DuckDB Memory?
MCP DuckDB Memory is a Model Context Protocol (MCP) server that memory server for claude that stores and retrieves knowledge graph data in duckdb, enhancing performance and query capabilities for conversations with persistent user information. It connects AI assistants to external tools and data sources through a standardized interface.
How do I install MCP DuckDB Memory?
Follow the installation instructions on the MCP DuckDB Memory GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.
Which AI clients work with MCP DuckDB Memory?
MCP DuckDB Memory works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.
Is MCP DuckDB Memory free to use?
Yes, MCP DuckDB Memory is open source and available under the MIT License license. You can use it freely in both personal and commercial projects.
MCP DuckDB Memory Alternatives — Similar Knowledge & Memory Servers
Looking for alternatives to MCP DuckDB Memory? Here are other popular knowledge & memory servers you can use with Claude, Cursor, and VS Code.
MemPalace
★ 52.6kA local AI memory system that stores all conversations verbatim and organizes them into navigable structures. It provides 19 MCP tools for AI assistants to search and retrieve past decisions, debugging sessions, and architecture debates automatically
Kratos
★ 25.7k🏛️ Memory System for AI Coding Tools - Never explain your codebase again. MCP server with perfect project isolation, 95.8% context accuracy, and the Four Pillars Framework.
Context Mode
★ 15.4kAn MCP server that preserves LLM context by intercepting large data outputs and returning only concise summaries or relevant sections. It enables efficient sandboxed code execution, file processing, and documentation indexing across multiple programm
Memu
★ 13.7kMemory for 24/7 proactive agents like OpenClaw.
MemOS
★ 9.3kMemOS (Memory Operating System) is a memory management operating system designed for AI applications. Its goal is: to enable your AI system to have long-term memory like a human, not only remembering what users have said but also actively invoking, u
Everos
★ 5.4kBuild, evaluate, and integrate long-term memory for self-evolving agents.
Browse More Knowledge & Memory MCP Servers
Explore all knowledge & memory servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.
Set Up MCP DuckDB Memory in Your Editor
Choose your AI client for step-by-step setup instructions.
Quick Config Preview
Add this to your claude_desktop_config.json or .cursor/mcp.json
Ready to use MCP DuckDB Memory?
Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.