EnderGamer6112/open-allscrape — reverse-engineered prompt

Build me a simple Python backend API for web scraping that is ready for LLM use. I want to give it one URL and get back clean content in text, markdown, and HTML, plus useful page details like title, description, author, date, word count, and source URL.

It should also let me search the web with a query, pick a limited number of results, scrape those pages automatically, and return everything as clean JSON. Please make it handle normal pages quickly, but also try a browser based fallback for dynamic pages when needed.

Use FastAPI and make sure the app can be run locally on port 8000. Add clear endpoints for /scrape and /search, with helpful error responses when a page fails. I also want the built in API docs to work so I can test requests in the browser. Keep the code clean and easy for a beginner to understand, and look up current docs online if you need to.