CyberScraper 2077

CyberScraper 2077 turns web pages into structured data through a conversational interface. It's for analysts and researchers who want to describe the information they need and export the results.

• Conversational extraction: Ask questions about a page and request selected data or the full page content through a Streamlit browser interface.
• Local and cloud models: Use local LLMs through Ollama, including Llama 3.1, or connect to OpenAI, Gemini and models exposed through a LiteLLM proxy.
• Data exports: Save extracted information as JSON, CSV, HTML, SQL or Excel, or send CSV data to Google Sheets through an authenticated Google connection.
• Tor access: Extract content from .onion sites through a Tor connection, with automatic routing for onion addresses.
• Browser handling: Use an existing local browser session and stealth settings to help with bot detection; interactive CAPTCHA handling works in native operation but isn't available in Docker.

The app is open source under the MIT license and runs locally on macOS or Linux, or in Docker. Windows users use Docker. Ollama can keep model processing on your own hardware; OpenAI and Gemini send extraction requests to external services and require API keys. Scraping websites still needs a network connection. Local model speed depends on your hardware, and extraction may need prompt adjustments or extra filters.