term

RAG (Retrieval-Augmented Generation)

Pattern of augmenting LLM responses with retrieved external data, typically from a vector database.

RAG addresses the LLM context-window limit by retrieving relevant documents from an external store and prepending them to the prompt. Self-hosted RAG setups in 2026 typically use Postgres with pgvector or a dedicated vector DB. ZeroClaw bundles a clean offline RAG implementation that works on any local document folder.

Related terms

ZeroClaw AI agent

Found a definition that's wrong, dated or could be sharper? Email us — we update with attribution unless you'd rather we didn't.