Extract clean, LLM-ready Markdown or plain text from up to 5 public webpage URLs. Removes page boilerplate and returns structured per-URL results for AI agents, RAG ingestion, research and summarization.