@markitdownjs/html

HTML to AST converter for MarkItDownJS. Uses Mozilla Readability for main-content extraction — well-suited for web pages, blog posts, and documentation sites.

Install

npm install @markitdownjs/html

Usage

import { MarkItDown } from "@markitdownjs/core";
import { HtmlConverter } from "@markitdownjs/html";

const parser = new MarkItDown();
parser.registerConverter(new HtmlConverter());

const result = await parser.convert({ source: htmlString, mimeType: "text/html" });
console.log(result.markdown);

Key Exports

| Export | Description | |---|---| | HtmlConverter | Converter plugin — register with MarkItDown |

What Gets Extracted

• Main article content via Mozilla Readability (strips nav, ads, footers)
• Headings, paragraphs, lists, tables, blockquotes, and code blocks
• <img> alt text and src preserved as ImageNode
• Page <title> and <meta> description captured as metadata

Options

parser.registerConverter(new HtmlConverter({
  useReadability: true,   // set false to parse full document without extraction
  baseUrl: "https://example.com", // used to resolve relative image/link URLs
}));

Accepted MIME Types

• text/html
• application/xhtml+xml

Part of the MarkItDownJS monorepo.