datawhalechina/all-in-rag ? reverse-engineered prompt

I want you to build a polished Chinese learning project about RAG, like an online handbook plus code examples, for people who want to learn how to build real AI question answering apps.

It should explain the basics first, then walk through data loading, text splitting, embeddings, vector search, hybrid retrieval, query rewriting, Text to SQL, formatted generation, and evaluation. Please also include a few practical demo projects, one focused on a normal document based RAG app, one with graph or knowledge graph style retrieval, and one chapter for multimodal image and text retrieval.

Make it feel like a real course site, with clear chapter pages, sample code, diagrams, example data, and a simple way to read everything online. Keep the style friendly and practical, and make sure the examples are runnable in Python. If you need to check current docs for libraries or tools, look them up online first.