#AISChallenge-Day 2 · AI Automation Society

#AISChallenge-Day 2

Just ran my first scrape with Claude Code + Firecrawl MCP — and it actually worked on the first try.

What I scraped:

Booking.com hotel search results for Dubai — 25 properties extracted, including hotel names, locations, ratings, review counts, property types, and direct booking URLs. All saved automatically to a clean CSV.

What surprised me:

I didn't have to tell Claude Code which Firecrawl tool to use or how to call it. I just dropped the URL, and it figured out: use firecrawl_scrape, add a waitFor for the JS-heavy page, parse the markdown, detect the page type (directory/listing), map the right columns, name the file correctly, and write the CSV.

Use case idea:

Hotel and short-term rental research for travel clients or property investors — scrape competitor listings across Booking.com, Airbnb, or local portals, track rating trends over time, and flag new properties entering a market. Run it on a schedule, and you've got a lightweight market intelligence feed without paying for an expensive data API.

Even connecting the MCP server is a win. Once it's wired up, the scraping part is almost trivially easy.

#AISChallenge

9 comments