Neutrally - AI Workspace MVP

Neutrally is a unified AI workspace where conversations and knowledge persist across different AI models and external developer tools.

Features

• 🤖 Multi-AI Provider Support: Switch seamlessly between OpenAI and Anthropic models mid-conversation
• 💬 Persistent Conversations: All conversations are saved and searchable
• 🧠 Shared Memory: Context and knowledge persist across different AI models
• 🔍 Smart Search: Search through conversation history by title, summary, or keywords
• 🎨 Dark Mode: Toggle between light and dark themes
• 🔐 Secure: API keys stored locally, conversations encrypted
• 🔌 VSCode Integration Ready: Memory API endpoints for external tool integration

Tech Stack

• Frontend: Next.js 14, React 18, TypeScript, TailwindCSS
• Backend: Next.js API Routes, Supabase PostgreSQL
• AI Providers: OpenAI, Anthropic
• Icons: Lucide React

Prerequisites

• Node.js 18+ and npm
• Supabase account and project
• OpenAI API key
• Anthropic API key

Installation

1. Clone the repository:

git clone https://github.com/yourusername/neutrally.git
cd neutrally

2. Install dependencies:

npm install

3. Set up environment variables:
   • Copy .env.example to .env.local
   • Fill in your Supabase and AI provider credentials:

NEXT_PUBLIC_SUPABASE_URL=your_supabase_project_url
NEXT_PUBLIC_SUPABASE_ANON_KEY=your_supabase_anon_key
SUPABASE_SERVICE_ROLE_KEY=your_supabase_service_role_key
OPENAI_API_KEY=your_openai_api_key
ANTHROPIC_API_KEY=your_anthropic_api_key

4. Set up Supabase database:

   • Go to your Supabase project SQL editor
   • Run the SQL from supabase_schema.sql

5. Run the development server:

npm run dev

6. Open http://localhost:3000 in your browser

Project Structure

neutrally/
├── app/                     # Next.js app directory
│   ├── api/                # API routes
│   │   ├── chat/          # Chat completion endpoints
│   │   ├── conversations/ # Conversation CRUD endpoints
│   │   └── memory/        # Memory system endpoints
│   ├── chat/              # Chat page
│   ├── settings/          # Settings page
│   └── page.tsx           # Home page
├── components/             # React components
│   ├── ui/                # Base UI components
│   ├── layout/            # Layout components
│   └── chat/              # Chat-specific components
├── services/              # Service layer
│   ├── aiService.ts       # AI provider management
│   ├── conversationService.ts
│   ├── memoryService.ts
│   └── summaryService.ts
├── lib/                   # Libraries and utilities
│   ├── supabase.ts        # Supabase client
│   └── aiProvider.ts      # AI provider implementations
└── types/                 # TypeScript type definitions

API Endpoints

Chat

• POST /api/chat - Generate AI response

Conversations

• GET /api/conversations - Get user's conversations
• POST /api/conversations - Create new conversation
• DELETE /api/conversations - Delete conversation
• GET /api/conversations/[id]/messages - Get conversation messages

Memory System (VSCode Integration)

• GET /api/memory/search - Search through conversation memory
• GET /api/memory/fetch - Fetch specific conversation details
• GET /api/memory/context - Get user context and recent conversations
• POST /api/memory/context - Update user context

Features in Detail

Conversation Summaries

• Automatically generated every 10 messages
• Extracts keywords, topics, and entities
• Updates user's global context

Model Switching

When switching between AI providers mid-conversation:

• Conversation summary is sent as context
• Last 10 messages are included
• User's global context is provided

Performance Optimizations

• Never sends full conversation history to models
• Uses summaries for context
• Limits message history to last 10 messages
• Implements efficient search with database indexes

Development

# Run development server
npm run dev

# Build for production
npm run build

# Start production server
npm start

# Run linting
npm run lint

Deployment

This project is ready to deploy on Vercel:

1. Push your code to GitHub
2. Connect your repository to Vercel
3. Add environment variables in Vercel dashboard
4. Deploy

Security Notes

• API keys are stored locally in the browser (Settings page)
• Supabase handles authentication and row-level security
• All database queries use prepared statements
• Conversations are isolated per user

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

License

MIT

Support

For issues and questions, please open an issue on GitHub.

Built with ❤️ using Next.js and AI