Scrapling adaptive scraping framework adds MCP and Agent skill support

Scrapling is an adaptive web scraping framework that handles everything from single requests to large-scale crawls. Its parser learns from site changes and relocates elements automatically when pages update; the fetcher bypasses anti-bot systems like Cloudflare Turnstile by default; and the crawler supports concurrency, multi-session crawling, pause/resume, automatic proxy rotation, adaptive speed based on target response times, and automatic backoff when blocked. The new MCP server lets AI agents such as Claude and Cursor scrape via one-off or session-based tools, covering plain HTTP requests, browser scraping, and stealth scraping that bypasses Cloudflare. Pages are first narrowed with CSS selectors and stripped of prompt injections, so agents read less, cost less, and can't be hijacked by hidden text. It also ships an installable Agent skill that helps coding agents write code matching the current API, plus a one-line page.markdown() that converts pages into clean LLM-ready Markdown.