🚀 Queen RAG Installer

Create production-ready RAG (Retrieval-Augmented Generation) applications with the Queen Claude ecosystem.

Features

• Full-Stack RAG Application: Next.js frontend + FastAPI backend
• Vector Database: ChromaDB for efficient document retrieval
• Assistant UI: Pre-configured chat interface with streaming
• OpenAI Integration: GPT-4 with vision capabilities
• Queen Claude Ecosystem: Specialized agents and commands for RAG development

Installation

npm install -g @kenkaiiii/queen-rag

Usage

queen-rag my-rag-app
cd my-rag-app

Project Structure

my-rag-app/
├── frontend/          # Next.js + Assistant UI
├── backend/           # FastAPI + ChromaDB
├── .claude/           # Queen Claude configuration
│   ├── agents/        # RAG-specialized agents
│   └── commands/      # Custom slash commands
└── CLAUDE.md          # RAG development guidelines

Specialized Agents

• rag_backend: FastAPI & ChromaDB specialist
• rag_frontend: React & Assistant UI specialist
• vector_specialist: Embeddings & search optimization
• debuggy: RAG system debugging
• cheeky_backend: Python validation
• cheeky_frontend: Frontend validation

Available Commands

• /start-rag: Get everything running quickly
• /check: Verify code quality and fix issues
• /commit: Create a git commit

Quick Start

Backend Setup

cd backend
python -m venv venv
source venv/bin/activate  # Windows: venv\Scripts\activate
pip install -r requirements.txt
cp .env.example .env
# Add your OpenAI API key to .env
python main.py

Frontend Setup

cd frontend
npm install
cp .env.example .env
npm run dev

Requirements

• Node.js 18+
• Python 3.10+
• OpenAI API key

Development Guidelines

The project includes strict RAG development guidelines in CLAUDE.md:

• KISS: Keep It Simple - use existing components
• YAGNI: Build only what's needed
• DRY: Don't repeat yourself
• SOLID: Single responsibility for agents

Quality Gates

Before committing:

• Backend: MyPy strict mode, Ruff checks
• Frontend: TypeScript strict, ESLint
• Vector ops: 1536 dimensions, cosine similarity
• Performance: <200ms search, <1s first token

License

MIT

Support

For issues and questions: https://github.com/kenkaiiii/queen-claude/issues