Hi everyone!! This is my build from day 2. I scraped an online bookstore (books.toscrape.com) with the Firecrawl MCP server. I pulled the title, price, rating, and availability for 60 books and saved it all into a clean CSV file. What surprised me: I just described what I wanted in plain English, and Claude Code picked the right Firecrawl tool by itself first map to see the site structure, then scrape to get the pages. I didn't have to know which tool to call. The MCP server gives Claude the tools, and it chooses the right one. For a business I could scrape competitor prices or product listings automatically for example track prices in a shop every day and drop them into a spreadsheet, so I always know where I stand. Or collect leads (companies and contact info) from online directories. On to Day 3!