# CyberScraper 2077

An AI-powered web scraper with a graphical interface that uses OpenAI, Gemini or local models to extract structured data from web pages, including through the Tor network.

CyberScraper 2077 is a web scraping tool that uses large language models, including OpenAI, Gemini and local models through Ollama, to understand and parse web content rather than relying on hand-written selectors. It has a Streamlit-based graphical interface and exports scraped data as JSON, CSV, HTML, SQL or Excel.

It can scrape .onion sites through the Tor network with automatic routing, includes a stealth mode intended to help the scraper avoid bot detection, runs asynchronously for speed, and caches content- and query-based results using an LRU cache and a custom dictionary to reduce redundant API calls.

It is written in Python and released under the MIT license.

Website: https://github.com/itsOwen/CyberScraper-2077