Mermaid Flowchart: How RAG Works

A Mermaid flowchart explaining Retrieval-Augmented Generation — from user question through retrieval to final answer.

Starter Prompt

Create a Mermaid flowchart that explains how RAG (Retrieval-Augmented Generation) works, from user question to final answer.

Expanded Prompt

Create a Mermaid flowchart that explains how RAG (Retrieval-Augmented Generation) works for a beginner-friendly AI course.

Goal

Build a clear flow diagram that shows how a user question is answered using both a large language model and external documents or knowledge sources.

Steps to Include

1. User asks a question
2. Question is converted into an embedding
3. Vector database searches for similar content
4. Relevant documents or chunks are retrieved
5. Decision point: Was relevant context found?

If Yes:

• Add retrieved context to the prompt
• Send prompt to the LLM
• LLM generates an answer
• Answer is returned to the user

If No:

• Ask the LLM to answer from general knowledge
• Optionally warn the user that no supporting documents were found
• Answer is returned to the user

Mermaid Requirements

• Use flowchart TD (top-down layout)
• Use clear, short node labels
• Label the Yes and No branches on the decision point
• Use classDef styling to visually distinguish the decision node if helpful

Design Goals

• Easy to read and follow
• Suitable for a beginner AI course or slide deck
• Clear about what happens before and after retrieval

Output Instructions

Return only the Mermaid code block. No explanation before or after.

Tips for Refining

• “Render this as an HTML artifact with the diagram displayed inline”
• “Add color to the Yes and No branches”
• “Add a step for re-ranking retrieved chunks before sending to the LLM”
• “Make a left-right (LR) layout version”