Loreto
Home/Glossary/RAG (Retrieval-Augmented Gene…
Glossary

RAG (Retrieval-Augmented Generation)

Quick answer

RAG (Retrieval-Augmented Generation) is a technique where an AI model retrieves relevant documents and uses them as context to generate answers grounded in real data instead of only its training.

RAG is great at factual lookup — "find passages that answer this." It struggles with questions about relationships, history, and causation, where a knowledge graph or a hybrid approach does better. In practice you route queries by type and add a fallback for when retrieval returns nothing relevant.

Related skills

RAG

RAG Query Routing

RAG

RAG Relevance Fallback

AI Architecture

Diagnosing RAG Failure Modes

Turn this into a skill your agent can run

Loreto packages any workflow or source into a production-ready Claude Code skill.

Generate a skill freeBrowse the marketplace