opencli-browser

Automate authenticated web scraping by driving a live Chrome browser session that maintains session persistence, handles login flows, and performs DOM manipulation to extract data. It applies a selector-first contract with numeric refs to handle dynamic UI re-renders automatically.

You can inspect network traffic and capture API responses directly during browser automation, bypassing complex DOM scraping for data-heavy tasks. It uses the Chrome DevTools Protocol to capture and inspect API responses for structured data extraction.

Handle dynamic DOM drift by using a selector-first contract with numeric refs that automatically adjusts to UI re-renders and structural changes. This approach ensures resilient interaction when driving live Chrome sessions for complex web tasks.

Yes, it works with legacy internal portals by driving a real Chrome instance that maintains specific browser state and session persistence. It automates multi-step login and data entry processes on complex web applications standard scraping tools cannot handle.

Live Chrome sessions are required for agentic web tasks involving authentication, multi-step flows, or real-time data inspection that standard HTML scraping tools cannot handle. It drives a real browser to click, type, select, and navigate through authenticated sessions reliably.

Limitations include relying on a live Chrome instance which requires more resources than lightweight HTTP scraping. It is not suited for simple static page extraction, but targets complex dynamic web applications needing session persistence and network traffic inspection.