Knowledge Graph Memory Server
I enhance the existing memory mcp server from the official mcp github, so big thanks and credits for creating this
Overview
The Knowledge Graph Memory Server is an enhanced implementation of a persistent memory system using a local knowledge graph, allowing Claude to remember user information across chats and learn from past errors through a lesson system.
To use the server, clone the repository, install dependencies, build the project, and configure it with the Cursor MCP client or Claude Desktop. Follow the setup instructions provided in the documentation.
- Persistent memory using a local knowledge graph - Ability to create and manage entities, relations, and observations - Lesson management for capturing knowledge about errors and solutions - API tools for creating, reading, updating, and deleting graph elements
- Storing user preferences and behaviors for personalized interactions
- Managing error patterns and solutions for software development
- Enhancing chatbots with memory capabilities for improved user experience
Add to your AI client
Use these steps to connect Knowledge Graph Memory Server in Cursor, Claude, VS Code, and other MCP-compatible apps. The same JSON appears in the Use with menu above for one-click copy.
Cursor
Add this to your .cursor/mcp.json file in your project root, then restart Cursor.
.cursor/mcp.json
{
"mcpServers": {
"memories-with-lessons-mcp-server-t1nker-1220": {
"command": "npx",
"args": [
"-y",
"@modelcontextprotocol/server-memories-with-lessons-mcp-server-t1nker-1220"
]
}
}
}Claude Desktop
Add this server entry to the mcpServers object in your Claude Desktop config, then restart the app.
~/Library/Application Support/Claude/claude_desktop_config.json (macOS) or %APPDATA%\Claude\claude_desktop_config.json (Windows)
{
"mcpServers": {
"memories-with-lessons-mcp-server-t1nker-1220": {
"command": "npx",
"args": [
"-y",
"@modelcontextprotocol/server-memories-with-lessons-mcp-server-t1nker-1220"
]
}
}
}Claude Code
Add this to your project's .mcp.json file. Claude Code will detect it automatically.
.mcp.json (project root)
{
"mcpServers": {
"memories-with-lessons-mcp-server-t1nker-1220": {
"command": "npx",
"args": [
"-y",
"@modelcontextprotocol/server-memories-with-lessons-mcp-server-t1nker-1220"
]
}
}
}VS Code (Copilot)
Add this to your .vscode/mcp.json file. Requires the GitHub Copilot extension with MCP support enabled.
.vscode/mcp.json
{
"servers": {
"memories-with-lessons-mcp-server-t1nker-1220": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"@modelcontextprotocol/server-memories-with-lessons-mcp-server-t1nker-1220"
]
}
}
}Windsurf
Add this to your Windsurf MCP config file, then restart Windsurf.
~/.codeium/windsurf/mcp_config.json
{
"mcpServers": {
"memories-with-lessons-mcp-server-t1nker-1220": {
"command": "npx",
"args": [
"-y",
"@modelcontextprotocol/server-memories-with-lessons-mcp-server-t1nker-1220"
]
}
}
}Cline
Open Cline settings, navigate to MCP Servers, and add this server configuration.
Cline MCP Settings (via UI)
{
"mcpServers": {
"memories-with-lessons-mcp-server-t1nker-1220": {
"command": "npx",
"args": [
"-y",
"@modelcontextprotocol/server-memories-with-lessons-mcp-server-t1nker-1220"
]
}
}
}FAQ
Can the server handle multiple users?
Yes! The server is designed to manage information for multiple users through unique entities.
Is there a limit to the number of entities or observations?
No, but performance may vary based on the size of the graph and the complexity of the data.
How can I contribute to the project?
Contributions are welcome! Please refer to the repository for guidelines on how to contribute.