Extract the main article from a web page with Mozilla Readability — drops nav, ads, sidebars and footer boilerplate. Send inline {html} or a public {url} to fetch (SSRF-checked). Returns the article title, byline, cleaned HTML content, plain textContent, excerpt, character length, language and site name. THE pre-processing step before sending a page to an LLM. Distinct from html-to-md (which renders Markdown). Max 5 MB.