Unix-style shell for navigating and querying web pages as a filesystem.

• Navigate URLs with cd, list page elements with ls, extract content with cat, and filter with grep—all operating on cached, locally-parsed page content
• Supports natural language intent inference; commands like "show me the first 5 links" or "what forms are on this page?" work alongside formal shell syntax
• Fully asynchronous design with instant prompt return; background tasks handle fetching, parsing, and eager prefetching of linked pages across configurable depth layers
• Maintains session state including command history, bookmarks, and cached pages in .websh/ directory; gracefully degrades if content isn't ready yet