imread

imread is a free article reader and text extractor. Paste any article URL and it returns the main text of the page without the navigation, ads, popups or scripts, ready to copy as plain text or Markdown. The output keeps real structure instead of flattening everything: headings, lists, links, footnotes and fenced code blocks survive, and data tables come out as tab-separated rows in plain text (they paste into a spreadsheet as columns) or pipe tables in Markdown. WeChat Official Account articles, which ship several megabytes of inline script, and SEC filings, with their wide financial tables, are both supported, as are pages in GBK and Shift_JIS. Paste plain text instead of a link and you can listen to it read aloud or translate it. A single-file PDF text reflow tool is included for text copied out of PDFs. For developers and AI agents there is a free REST API: GET /api/extract?url=... returns JSON with title, byline, word count, headings, tables and the article text, and format=markdown or format=text returns just the article. An MCP server exposes the same extractor as a tool. Normal use needs no account and no API key.