bx-ai: RAG (Retrieval-Augmented Generation)

RAG enhances AI responses by grounding them in your own documents, reducing hallucinations and keeping answers current without model retraining.

RAG Workflow

Documents → Load → Chunk → Embed → Vector DB
                                       ↓
User Query → Embed → Vector Search → Retrieve → Inject into Context → AI

Quick Start: Complete RAG System

// 1. Create vector memory (ChromaDB — production)
vectorMemory = aiMemory( "chroma", config: {
    collection       : "knowledge_base",
    embeddingProvider: "openai",