CyberYozh launches free open-source Yozh Scraper for AI and data teams

A new tool for data collection

CyberYozh App, a company that provides proxy and online infrastructure, has released a new toolkit called Yozh Scraper. This free, open-source software is designed to help AI engineers and data teams gather publicly available information from the web. It aims to solve the problem of teams having to build complex scraping infrastructure from the ground up.

The toolkit is built on Playwright, which is a technology used for browser automation. By using this foundation, the tool can handle both individual page scraping and full-site crawling. It is intended to be a flexible environment where developers can manage various web data extraction projects more efficiently.

As reported by Yahoo Finance UK, the release is intended to support the growing demand for reliable web data in the fields of analytics and digital business. Because it is open-source, it allows technical teams to maintain greater control over their configurations and the information they collect.

How the scraper functions

Yozh Scraper is divided into two primary components to handle different types of data needs. The Open Scraper component is used to collect information from specific web pages using predefined presets. This is useful for targeted data gathering where the user knows exactly which page they need to access.

The Open Crawler component provides the ability to navigate and collect data across entire websites. This allows for broader data mining projects where a team might need to index a whole domain. Together, these two tools give developers a versatile set of options for different scales of data collection.

The toolkit also includes several advanced technical features to ensure reliability. These include persistent sessions, cookie handling, and proxy configuration. It can output data in various formats, such as HTML, specific extracted data fields, or even screenshots of the pages.

Integration with AI workflows

One of the most significant features for modern developers is the toolkit's support for AI-powered workflows. It includes a REST API and a Model Context Protocol (MCP) endpoint. These features allow compatible AI agents and external applications to request web data directly from the scraper.

Developers also have the option to enable large language model-assisted parsing. This means they can use their own model provider API keys to help the system interpret and organize the data it finds. This creates a bridge between raw web data and the structured information that AI models need to function.

To make the process faster, the toolkit comes with ready-made presets for several major platforms. These include popular sites such as Amazon, Google, eBay, Walmart, YouTube, and LinkedIn. By using these presets, teams can skip the initial setup phase for some of the most common data sources.

Infrastructure and cost control

The self-hosted nature of Yozh Scraper is a key differentiator from many managed scraping platforms. Many commercial services charge a fee for every request made to a website. In contrast, this open-source toolkit allows users to run their own workloads and manage their own infrastructure costs independently.

Users can choose to run the software locally or on a server using Docker, which is a platform for containerizing applications. This setup gives technical teams more autonomy over where their data is processed. It is particularly useful for organizations that have strict requirements regarding how they handle data flow.

While the toolkit is free, it does offer integration with CyberYozh proxy infrastructure. This is optional and allows users to access rotating residential, static residential, LTE/5G mobile, and datacenter proxies. These proxies help with geographic targeting by country, region, or city, though they are not required for projects that work on a direct connection.

Who this affects

This release primarily targets AI engineers and data teams who are looking for more control over their data pipelines. These professionals often need to feed large amounts of web data into models for training or real-time analysis. Having a self-hosted tool can reduce the overhead of managing third-party subscriptions.

Developers who work on digital business operations will also find the tool useful for automated workflows. The ability to handle dynamic websites and location-specific data collection is a common requirement for these roles. The toolkit addresses these needs by providing a robust set of tools in one package.

By providing an open-source option, CyberYozh is making it easier for smaller teams to access high-quality scraping capabilities. It levels the playing field for those who may not have the budget for expensive managed services but still need to perform complex web data extraction tasks.