Does it preserve readable (ad stripped) versions of pages I collect from falling to link rot, and does it automatically convert everything into content type ada

Does it preserve readable (ad stripped) versions of pages I collect from falling to link rot, and does it automatically convert everything into content type adapative embeddings for citable chat querying with GPT-4 / user’s own personal LLM?