Agent01 - Local AI Chat Platform

A complete local AI chat platform built with Next.js, Ollama, and ChromaDB. Run entirely offline with no cloud dependencies.

IDE

VSCode, Github Copilot Agent with Claude Sonnet 4

🎯 Core Features

• 🔒 100% Local - Everything runs locally: AI models, vector database, and web interface
• 🤖 Multiple AI Models - Use any Ollama model with automatic tool support detection
• � Vector Database - ChromaDB integration for document search and context augmentation
• 🔧 Built-in Tools - Web Search, Dummy Document Summarizer, Add more...
• 📎 File Support - Upload images, PDFs with multimodal processing
• ⚡ Streaming Responses - Real-time AI responses

🚀 Quick Start

Prerequisites

• Node.js 18+ - Download here
• Ollama - Install from ollama.ai
• Python 3.8+ - For ChromaDB (optional)

1. Setup the Platform

# Clone and install
git clone https://github.com/dchpro87/agent01.git
cd agent01
npm install

# Install required AI models
ollama pull llama3.2:3b          # Primary chat model
ollama pull nomic-embed-text     # For vector embeddings

2. Start Services

# Terminal 1: Start Ollama
ollama serve

# Terminal 2: Start ChromaDB (optional but recommended)
pip install chromadb
chroma run --host localhost --port 8000

# Terminal 3: Start the platform
npm run dev

3. Access the Platform

Open http://localhost:3000 in your browser

💡 How to Use

Basic Chat

• Type messages and get AI responses with real-time streaming
• Switch models using the 🤖 icon for different capabilities
• Use 🎭 icon to select different AI personalities

Document Context (ChromaDB)

• Click 🗄️ icon to manage document collections
• Add collections to chat context for intelligent Q&A
• Upload your own documents (PDFs, etc.) to create knowledge bases

Tools & Features

• File Attachments: Click 📎 to upload images, documents
• AI Tools: Ask for time, calculations, or web searches (with API key)
• Model Config: Adjust temperature, tokens, and other parameters

🛠️ Tech Stack

• Frontend: Next.js 15, React 19, TypeScript, Tailwind CSS
• AI: Ollama (local models), Vercel AI SDK v4
• Vector DB: ChromaDB with nomic-embed-text embeddings
• Optional: SerpAPI for web search

📁 Project Structure

src/
├── app/api/          # API routes for chat, ChromaDB, health
├── components/       # React components for UI
├── constraints/      # Configuration and constants
├── lib/             # Core logic (AI, ChromaDB, tools)
├── types/           # TypeScript type definitions
└── utils/           # Helper functions

⚙️ Configuration

Edit src/constraints/app-config.ts to customize:

export const APP_CONFIG = {
  ollama: {
    baseURL: "http://localhost:11434",  // Ollama server
    model: "llama3.2:3b",               // Default model
    temperature: 0.7,                   // Response creativity
  },
  // ... other settings
};

� Development

npm run dev          # Start development server
npm run build        # Build for production  
npm run lint         # Run code quality checks

🔧 Adding Features

New AI Tools

Create in src/lib/tools/ and export from src/lib/tools/index.ts

New AI Personalities

Add to PREDEFINED_PROMPTS in src/constraints/predefined-system-prompts.ts

Custom Models

Any Ollama model works - just add it to the model database in src/constraints/model-databse.ts

📄 License

MIT License

---

Built for local AI development by dchpro87