Kimurai is an open source web scraping framework written in Ruby that simplifies the process of building automated data extraction tools. It provides a clean domain-specific language that allows developers to define scraping logic and data schemas with minimal boilerplate code. Kimurai can use AI-assisted extraction to identify where data resides in HTML pages, automatically generating selectors that are cached for future use so subsequent scraping runs operate with pure Ruby performance. Kimurai supports scraping both static and JavaScript-rendered websites by working with multiple engines, including headless browsers and simple HTTP-based approaches. Developers can also interact with pages using browser automation features such as form filling, clicking elements, or navigating through dynamic content. It includes tools for scheduling, parallel scraping, and structured data output, making it suitable for building reliable large-scale crawlers.
Features
- AI-assisted data extraction that generates reusable XPath selectors
- Supports headless Chrome, headless Firefox, and HTTP-based scraping engines
- Ability to scrape JavaScript-rendered pages and dynamic content
- Built-in helpers for exporting scraped data to JSON, JSON Lines, or CSV
- Parallel crawling and scheduling support for large scraping tasks
- Interactive console and debugging tools for developing scraping spiders