Memory
A basic implementation of persistent memory using a local knowledge graph. This lets Claude remember information about the user across chats.
New to MCP? Copy the config below, then follow the Claude Desktop / Cursor steps.
Install guide → npx -y @modelcontextprotocol/server-memory Installation Guide
Detailed Documentation
What is Memory?
A basic implementation of persistent memory using a local knowledge graph. This lets Claude remember information about the user across chats.
Memory is listed in the WorkflowAI MCP directory so agent builders can discover, install, and configure it quickly for Claude Desktop, Cursor, and other MCP-compatible clients.
Why use it
- Extend AI agents with Memory capabilities without writing a custom integration from scratch
- Standard Model Context Protocol interface — works across MCP-compatible clients
- Open source on GitHub — review code before production use
Requirements
- An MCP-compatible client (Claude Desktop, Cursor, Windsurf, etc.)
- Node.js 18+ (for
npx) - Any API keys or credentials listed in the upstream README
Install & configure
Official reference server (memory)
npx -y @modelcontextprotocol/server-memoryPublished by the Model Context Protocol project. Add the config JSON to Claude Desktop / Cursor and restart.
Example MCP config
{
"mcpServers": {
"memory": {
"command": "npx",
"args": [
"-y",
"@modelcontextprotocol/server-memory"
],
"env": {}
}
}
}Source & attribution
| Field | Value |
|---|---|
| Repository | modelcontextprotocol/servers |
| Stars | 88.371 |
| Language | TypeScript |
| License | See repository |
| Topics | No topics listed |
Attribution: Metadata and install hints are derived from the public GitHub repository. WorkflowAI does not claim ownership of this server. All rights remain with the original authors.
Related
Configuration
{
"mcpServers": {
"memory": {
"command": "npx",
"args": [
"-y",
"@modelcontextprotocol/server-memory"
],
"env": {}
}
}
}