Extracts structured data from complex websites using a robust, three-phase Playwright automation workflow.
This skill provides a standardized framework for building resilient web scrapers with Playwright within the Claude Code environment. It guides users through a rigorous 3-phase process—Login, Index, and Show—ensuring that authentication is handled persistently, site hierarchies are accurately mapped, and data is extracted using stable semantic selectors. By emphasizing incremental script development over ephemeral exploration, this skill enables the creation of scrapers that are reproducible, recoverable from failures, and capable of handling dynamic modern web frameworks like React and HeadlessUI.
Key Features
010 GitHub stars
02Standardized 3-phase sequential workflow: Login, Index, and Show
03Robust authentication persistence using storage state patterns
04Semantic selector design for stability against minor site updates
05Incremental script building with built-in verification and screenshot checkpoints
06Advanced handling for dynamic SPA components like tabs and accordions
Use Cases
01Scraping high-volume product catalogs or documentation from complex SPAs
02Automating data extraction from authenticated dashboards or paywalled content
03Building long-running, recoverable scrapers for paginated or nested data structures
