Firecrawl

A solid web scraping connector that covers the main Firecrawl API endpoints: single page scraping, full site crawling, URL mapping, web search, and AI-powered data extraction. The crawl operation is async, so you kick it off, get a job ID back, then poll for results. The extract endpoint is interesting because it uses AI to pull structured data based on a prompt or JSON schema, which beats writing custom parsers. All requests go through curl with Bearer token auth. You'll mostly use scrape for quick grabs and crawl when you need an entire site's worth of pages. The skill documentation shows practical examples for each endpoint with actual JSON payloads, which helps when you're trying to remember the parameter names.