# Universal RAG MCP

> 🧠 Intelligent cross-platform memory system for AI assistants

Give your AI assistants **persistent, searchable memory** that works across Claude, ChatGPT, Gemini, and more.

[![npm version](https://badge.fury.io/js/@sid7vish%2Funiversal-rag-mcp.svg)](https://www.npmjs.com/package/@sid7vish/universal-rag-mcp)
[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)

## ✨ Features

- 🧠 **Smart Memory** - Intelligent chunking filters noise, keeps only important information
- 🔍 **Semantic Search** - Find information by meaning, not just keywords
- 🚫 **Auto-Deduplication** - Tracks mention count instead of saving duplicates
- 🌐 **Cross-Platform** - Same memory across Claude, ChatGPT, Gemini, Cursor, Kiro
- ⚡ **Fast** - Sub-100ms searches with in-memory caching
- 🎯 **Accurate** - Multi-question support with parallel searches
- 🔒 **Your Data** - Stored in your Firebase/Pinecone accounts

## Features

- **Cross-platform**: Same memory in Claude Desktop, ChatGPT, Gemini, and more
- **Your data**: You control it - stored in your Firebase/Pinecone accounts
- **Zero config**: 5-minute setup wizard handles everything
- **Smart search**: Semantic search with automatic reranking
- **Fast**: In-memory cache + hot/warm/cold storage tiers

## 🚀 Quick Start

```bash
# 1. Install
npm install -g @sid7vish/universal-rag-mcp

# 2. Setup (5 minutes)
universal-rag-mcp init

# 3. Add MCP config to your AI platform (shown at end of setup)

# 4. Restart your AI platform and test!
```

**That's it!** Your AI now has persistent memory.

## 📖 Documentation

**[→ Read the Complete Guide](./GUIDE.md)**

Everything you need:
- Detailed setup instructions
- API keys walkthrough
- Platform configuration
- Troubleshooting
- Advanced usage

## 💬 Example Usage

```
You: Remember that I love TypeScript and I'm building SLAM v3
AI: Got it!

You: What am I working on?
AI: You're building SLAM v3 and you love TypeScript!
```

## 🏗️ Architecture

```
AI Platform (Claude/ChatGPT/Gemini)
    ↓ MCP Protocol
universal-rag-mcp
    ↓
Firebase (data) + Pinecone (vectors) + Voyage AI (embeddings)
```

## 🔑 What You Need

**4 Free API Keys** (setup wizard guides you):

1. **Firebase** - Database storage (free: 1GB)
2. **Pinecone** - Vector search (free: 5M vectors)
3. **Voyage AI** - Primary embeddings (free: 10M tokens)
4. **Cohere** - Fallback embeddings (free: 1K calls/month)

**Total setup time:** 5 minutes  
**Monthly cost:** $0 (free tiers cover most users)

## 📚 Commands

```bash
universal-rag-mcp init      # Setup wizard
universal-rag-mcp status    # Check configuration
universal-rag-mcp config    # Show config location
```

## 🤝 Contributing

Contributions welcome! Open an issue or PR.

## 📄 License

MIT License - See [LICENSE](./LICENSE)

## 🔗 Links

- **NPM Package:** https://www.npmjs.com/package/@sid7vish/universal-rag-mcp
- **GitHub Repo:** https://github.com/Sid7on1/universal-memory-mcp
- **Complete Guide:** [GUIDE.md](./GUIDE.md)

---

**Made with ❤️ by [Siddharth Vishwanath](https://github.com/Sid7on1)**