<<<<<<< HEAD

AssisantAI - Master Hub

Your Personal "MegaMan" AI Assistant - A master hub for all Niche AI Assistants that run locally on your computer. Think of it like MegaMan from the cartoon with LAN - your personal AI assistant who lives in your devices, knows you intimately, and can jump between specialized applications while maintaining the same personality, memory, and understanding of your goals.

Read the Vision Document to understand the full philosophy and goals of AssisantAI.

Unified AI Assistant Infrastructure - A master project that connects all AI assistant applications under a single agent infrastructure, ensuring users have the same account, agent, and information across all apps.

Architecture

AssisantAI/
├── apps/                    # Child AI assistant applications
│   ├── personalai/         # PersonalAI (Backend + Frontend)
│   └── mvpassistant/       # MVP Assistant (Backend + Frontend)
├── packages/               # Shared packages
│   ├── auth/               # Unified authentication
│   ├── agent/              # Shared agent infrastructure
│   └── port-manager/       # Port management system
├── hub/                     # Central hub dashboard (port 4200)
├── middleware/              # API middleware connecting all apps (port 4199)
├── config/                  # Configuration files (ports, apps)
├── install.sh               # Complete installation script
├── start.sh                 # Start all services
└── stop.sh                  # Stop all services

Features

• Unified Authentication: Single sign-on across all apps
• Shared Agent: Same AI agent personality and memory across apps
• Full Conversation History: Every message automatically includes complete conversation history for context
• Organic Learning System: Learns about users naturally through conversation to provide better service
• Cross-App Data: User data and conversation history shared
• Central Hub Dashboard: Visual interface to manage all apps (port 4200)
• Automatic Port Management: Apps automatically assigned ports (4199, 4200, 4201, 4202...)
• Middleware Layer: Connects all apps to shared backend (port 4199)
• Complete Bundle: Everything included - just download, install, and run

Quick Start

Installation

One command installs everything:

./install.sh

This single command will:

• ✅ Install all system dependencies (Homebrew, Python, Node.js, Ollama)
• ✅ Download and install the AI model (llama3:latest)
• ✅ Install all project dependencies (root, packages, middleware, hub)
• ✅ Install PersonalAI (Backend + Frontend)
• ✅ Install MVP Assistant (Backend + Frontend)
• ✅ Create all configuration files
• ✅ Build all packages
• ✅ Verify all dependencies are installed

⏱️ Takes 15-30 minutes (mostly downloading packages and the AI model)

Starting the Application

Start all services:

./start.sh

This starts:

• Middleware (port 4199)
• Central Hub (port 4200)
• MVP Assistant Backend (port 4201)
• PersonalAI Backend (port 4202)
• PersonalAI Frontend (port 4203)
• MVP Assistant Frontend (port 4204)

Access Points

After starting, access:

• 🎛️ Central Hub: http://localhost:4200
• 🔌 Middleware API: http://localhost:4199
• 🤖 PersonalAI Backend: http://localhost:4202
• 🎨 PersonalAI Frontend: http://localhost:4203
• 🚀 MVP Assistant Backend: http://localhost:4201
• 💼 MVP Assistant Frontend: http://localhost:4204

Stopping Services

Stop all services:

./stop.sh

Port Assignment

Ports are automatically assigned and managed:

• 4199: Middleware (API gateway)
• 4200: Central Hub (dashboard)
• 4201: MVP Assistant Backend
• 4202: PersonalAI Backend
• 4203: PersonalAI Frontend
• 4204: MVP Assistant Frontend

See Port Management Guide for details.

Child Apps

This bundle includes two child applications that are automatically installed:

PersonalAI

• Backend: Port 4202 - Comprehensive local AI assistant with chat, image/video generation, song writing, and 20+ automation skills
• Frontend: Port 4203 - Web interface for PersonalAI
• Description: Serves as the base backend for AssistantAI ecosystem, providing authentication, agent infrastructure, and core services

MVP Assistant

• Backend: Port 4201 - Dynamically generates custom GUI applications to solve user problems
• Frontend: Port 4204 - Web interface for MVP Assistant
• Description: Problem-driven app generator that creates complete applications on-demand

Niche Assistants

AssisantAI is designed to host many specialized AI assistants, each an expert in their domain. See Niche Assistants Guide for the full list.

Planned Assistants Include:

• 🎨 3D Printing Assistant: English-to-CAD-to-STL converter
• 🎨 Logo Designer: AI-powered logo creation
• 🎵 Song Creator: Enhanced music generation
• 💻 Code Assistant: Project-aware coding help
• ✍️ Writing Assistant: Style-consistent content creation
• 🍳 Recipe Assistant: Personalized meal planning
• 💪 Fitness Assistant: Goal-tracking workouts
• 💰 Finance Assistant: Budget and goal management
• 📚 Learning Assistant: Personalized education
• And many more...

The Magic: All assistants share the same agent personality and knowledge, so your assistant remembers everything across all apps.

Development

Project Structure

• apps/ - Child applications (PersonalAI, MVP Assistant)
• packages/ - Shared TypeScript packages
• hub/ - Central dashboard (Next.js)
• middleware/ - API middleware (Express/TypeScript)
• config/ - Configuration files (ports.json, apps.json)

Adding New Apps

1. Create app directory in apps/
2. Add app configuration to config/apps.json
3. Ports are automatically assigned via config/ports.json
4. Run ./install.sh to install dependencies
5. App will be available via ./start.sh

Documentation

• Architecture - System architecture and design
• Port Management - How ports are assigned
• Integration Guide - How to integrate apps
• Vision - Project vision and philosophy

Repository

This project is hosted at: https://github.com/austinbrady/Assist

License

See LICENSE file for details.

Personal AI

A completely local AI service for chat, image generation, video processing, and song writing. Everything runs on your computer - no cloud, no restrictions.

🚀 Quick Start

First time setup:

1. Run the installer:

   ./INSTALL.sh

   This will install all prerequisites (Python, Node.js, Ollama) and dependencies.

2. Start the application:

   ./START.sh

The application will:

• ✅ Open a GUI window with toggle switch
• ✅ Display localhost and network IP addresses
• ✅ Start backend server (in a Terminal window) when toggled ON
• ✅ Start frontend server (in a Terminal window) when toggled ON
• ✅ Show blue neon update indicator when updates are available

📋 Prerequisites

Before running, make sure you have:

1. Python 3.8+ - Download
2. Node.js 16+ - Download
3. Ollama - Install Guide

Installing Ollama

macOS:

brew install ollama

Linux:

curl -fsSL https://ollama.ai/install.sh | sh

Windows: Download from ollama.ai/download

🎯 First Time Setup

1. Install Prerequisites:

   • Python 3.8+ (Download)
   • Node.js 16+ (Download)
   • Ollama (Install Guide)

2. Start Ollama:

   ollama serve

3. Install the AI model:

   ollama pull llama3:latest

4. Run Personal AI:

   ./START.sh

The first time you run it, it will automatically install all dependencies. This may take a few minutes.

When it starts, you'll see a display showing:

• Local: http://localhost:7777
• Network: http://YOUR_IP:7777

🌐 Access

Once running:

• Local: http://localhost:7777
• Network: http://YOUR_IP:7777 (shown in terminal)

🎮 Usage

./INSTALL.sh   # Install dependencies (first time only)
./START.sh     # Launch the GUI application

The GUI provides:

• Toggle Switch: Turn servers ON/OFF
• Status Display: Real-time backend/frontend status
• IP Addresses: Click to open in browser
• Update Indicator: Blue neon light when updates available

✨ Features

• Chat: Unrestricted AI conversations with personal assistants
• Image Generation: Create images from text
• Image Editing: Edit images with natural language
• Video Generation: Generate videos from prompts
• Song Writing: Write songs with "For fans of" inspiration
• User Accounts: Personal login with AI assistant selection
• Mobile Support: Responsive design for phones and tablets
• 100% Local: All processing on your machine

👥 User Accounts

Each user gets:

• Personal login (username + password)
• Choice of 7 biblical AI assistants
• Personal conversation history
• Private file storage

🤖 AI Assistants

Choose from 7 biblical archangels, each with unique personalities:

• Michael - Noble protector and leader
• Gabriel - Clear communicator and messenger
• Raphael - Compassionate healer
• Uriel - Wise scholar and guide
• Ariel - Bold warrior companion
• Jophiel - Creative artist
• Chamuel - Loving friend

📚 Documentation

Everything is self-contained. Run ./INSTALL.sh once, then ./START.sh to launch the GUI.

🔒 Privacy

• 100% LOCAL PROCESSING - No data leaves your computer
• All AI processing via local Ollama
• All files stored locally
• Zero internet connectivity for photos, videos, or user data
• No external API calls

🔐 Security

⚠️ IMPORTANT: Before pushing to GitHub, ensure all API keys and credentials are removed. See SECURITY.md for details.

🛠️ Troubleshooting

Application won't start:

• Make sure Python 3.8+ and Node.js 16+ are installed
• Check that Ollama is running: ollama serve
• Verify the model is installed: ollama list

Port already in use:

• Stop other applications using ports 8000 or 7777
• Or use the toggle switch in the GUI to restart servers

Installation fails:

• Make sure you have internet connection (for downloading dependencies)
• Check that you have write permissions in the directory
• Verify prerequisites are installed: python3 --version, node --version, ollama --version

📝 License

Private use only.

---

Ready? Just run ./START.sh and enjoy!

4ca1946f8db7d30d7fa1413e95ed00a23d87dab8