Scraping software

Python toolkit with request and browser modes. The current 30-site test uses its browser workflow.

Let an AI agent control a browser to complete a task.

Use a modified Firefox browser built for automation and anti-detection.

Control Chrome directly through its debugging interface.

A browser built to reduce automation detection. We have not tested it publicly.

Crawl pages with queues, retries and browser options. The current test measures one-page CheerioCrawler HTTP downloads, not crawling scale.

Make HTTP requests with browser connection profiles. The current 30-site pass uses direct Chrome150 mode without JavaScript.

Write short Python scripts to click and read browser pages.

Fetch pages with synchronous or asynchronous Python code.

A lightweight browser engine for rendering pages. Stealth is not its focus.

Control Chromium with Playwright-style code and stealth patches.

Control a browser for JavaScript pages and clicks. The current 30-site test uses plain headless Chrome with stock driver defaults.

Control Chrome from JavaScript. The current 30-site pass uses plain headless Chrome, stock launch defaults and one navigation per URL.

Control Chrome from Python through its debugging connection.

Fetch HTML in Python with a simple request. It does not run page JavaScript.

Fetch and parse pages in Python. Its browser modes need separate tests.

Crawl sites in Python with queues, parsing and pipelines. The current test measures single-page HTTP downloads, not crawling scale.

Python browser automation with helpers and UC mode. The current 30-site pass uses standard headless Chrome with UC off.

Run multi-step browser tasks with an AI system.

Crawl many URLs with a Rust-based tool. Crawl coverage has not been tested here.

Give a browser higher-level tasks with help from an AI model.

Fetch pages with browser-like network fingerprints, without launching a browser.