@mdgate/document

The shared document model and GitHub-Flavored Markdown renderer.

Official converters that can express structure build a Document here, then call documentToMarkdown. A table-escaping fix in this package applies to every format on that path.

Works in Node.js, Cloudflare Workers, Edge runtimes, and browsers.

npm install @mdgate/document

Why @mdgate/document

Without a shared model, every parser emits its own Markdown dialect: different escaping, different tables, different footnotes.

This package is that dialect:

• headings, paragraphs, lists, tables, quotes, code
• bold, italic, strikethrough, links, inline code
• footnotes and endnotes
• heading anchors
• merged cells

Use it when you write a converter and want the same Markdown as @mdgate/docx and @mdgate/pdf.

Build and render

import type { Converter } from '@mdgate/core';
import { documentToMarkdown, emptyDocument, heading, plain } from '@mdgate/document';

export function myFormat(): Converter {
  return {
    id: 'my-format',
    sniff(bytes) {
      return looksLikeMyFormat(bytes) ? 2 : 0;
    },
    convert(bytes) {
      const doc = emptyDocument();
      doc.blocks.push(heading(1, [plain('Title')]));
      return { markdown: documentToMarkdown(doc) };
    },
  };
}

You do not need this package to call toMarkdown. You need it to implement a converter that should match the official dialect.

Need a complete reader?

npm install @mdgate/converters

import { toMarkdown } from '@mdgate/converters';

const markdown = await toMarkdown(bytes, {
  path: filename,
});

@mdgate/document is part of the open-source mdgate/converters project.

For AI agents, the same converter architecture can be used to extend read_file from text files to real-world document formats.

License

MIT