Build-a-Prompt Series: Challenge 101.1 #3

Optimizing the RAG Loop & Token Economy: AI reasoning engines operate using a Retrieval-Augmented Generation (RAG) loop[cite: 2]. When a prompt lacks explicit context or structure, the model expends excess tokens wandering through unstructured text to find simple facts, which wastes computational money[cite: 1, 2].

Inspect the two prompt layouts below designed to query a dense influencer fashion guide. Select the single prompt structure that minimizes token waste and helps the RAG loop find the correct answer instantly.