Supermemory

Active
GitHub TypeScript MIT

Description

An extremely fast and scalable memory engine for the AI era. Provides a unified Memory API for AI applications with large-scale knowledge storage and efficient retrieval.

Key Features

  • Conversational Memory Extraction — Automatically extracts facts from conversations handling temporal changes and contradictions
  • User Profile Building — Auto-maintains user preferences, projects, and recent activity context summaries
  • Hybrid Search — RAG and memory search unified, retrieving knowledge docs and personalized context together
  • Connector Ecosystem — Auto-sync with Google Drive, Gmail, Notion, OneDrive, and GitHub
  • Multi-Modal Extractors — Handles PDFs, image OCR, video transcription, and code AST-aware chunking
  • MCP Integration — Provides persistent memory to AI tools like Claude and Cursor via MCP server

Use Cases

💡 Providing cross-conversation persistent memory for AI assistants to remember user preferences and project context
💡 Building a personal knowledge brain that auto-organizes and retrieves key information from past conversations
💡 Injecting real-time user profiles into AI agents for personalized responses
💡 Connecting external data sources (Gmail, Notion, etc.) to build a unified knowledge base
💡 Adding project memory to coding tools like Cursor/Claude Code via MCP protocol

Quick Start

npm install supermemory

import Supermemory from 'supermemory';
const client = new Supermemory();

// Store conversation memory
await client.add({
  content: 'User prefers functional TypeScript patterns',
  containerTag: 'user_123',
});

// Get user profile and relevant memories
const { profile, searchResults } = await client.profile({
  containerTag: 'user_123',
  q: 'What programming style does the user prefer?',
});

console.log(profile.static);  // Long-term facts
console.log(profile.dynamic); // Recent context

Related Projects