Repolode
Submit
alchaincyf

alchaincyf/huashu-md-html

A Claude Code-style skill that converts files and web pages to Markdown, turns Markdown into styled HTML or print-ready DOCX, and archives HTML back to Markdown.

View on GitHub
Stars
920
Forks
98
Open issues
6
Last commit
8 days ago

What is it?

huashu-md-html is a set of conversion scripts packaged as an agent skill. It converts PDF, DOCX, PPTX, XLSX, EPUB, images, audio, YouTube videos and web URLs into Markdown using markitdown. It can also render Markdown into HTML with four bundled single-file themes (article, report, reading, interactive) built on Pandoc.

In the reverse direction, it extracts the main body of a published HTML page or URL back into Markdown using html-to-markdown and trafilatura, removing navigation and sidebars. A separate script builds a DOCX from Markdown with book-style presets, including a cover, table of contents, headers and footers, chapter page breaks, and 32-open or A4 page sizes.

The skill installs with npx skills add and is designed to work across agents such as Claude Code and Cursor. Its scripts check dependencies on startup and report missing packages instead of failing silently.

Key features

  • 01Converts PDF, Office files, EPUB, images, audio, YouTube and web URLs to Markdown.
  • 02Renders Markdown into HTML with four self-contained themes that need no external CDN.
  • 03Extracts article body from web pages and strips navigation, sidebars and ads.
  • 04Builds book-style DOCX with cover, table of contents, headers, footers and chapter breaks.
  • 05Runs as a cross-agent skill installable with npx skills add.

Use cases

  • Turning a PDF whitepaper into a readable HTML page with the report theme.
  • Archiving a published blog post back into Markdown source.
  • Producing a formatted manuscript DOCX for an editor or submission system.
  • Converting a web doc page into structured Markdown for an AI assistant.

Quick start

  1. 1

    Install the skill with skills.sh

    npx skills add alchaincyf/huashu-md-html
  2. 2

    Install Python dependencies for markitdown

    python3 -m pip install 'markitdown[all]'
  3. 3

    Install pandoc (macOS example from the README)

    brew install pandoc
  4. 4

    Convert a markdown file to a styled HTML page

    python3 scripts/md_to_html.py article.md --theme article -o article.html

Things to know

  • Requires Python 3 plus several external tools such as pandoc, and some features need additional pip packages.
  • Designed to run as an agent skill; there is no standalone GUI.

Topics

#markdown#pandoc#html#docx#pdf-conversion#web-scraping#claude-code#document-conversion

FAQ

What is huashu-md-html?
A Claude Code-style skill that converts files and web pages to Markdown, turns Markdown into styled HTML or print-ready DOCX, and archives HTML back to Markdown. huashu-md-html is a set of conversion scripts packaged as an agent skill. It converts PDF, DOCX, PPTX, XLSX, EPUB, images, audio, YouTube videos and web URLs into Markdown using markitdown. It can also render Markdown into HTML with four bundled single-file themes (article, report, reading, interactive) built on Pandoc.
Is huashu-md-html free to use? What license does it have?
huashu-md-html is open source under the MIT license. Check the license terms before using it in commercial products.
Is huashu-md-html still maintained?
Actively maintained. The last commit to huashu-md-html was on Oct 13, 2026.
What can I use huashu-md-html for?
Turning a PDF whitepaper into a readable HTML page with the report theme. Archiving a published blog post back into Markdown source. Producing a formatted manuscript DOCX for an editor or submission system. Converting a web doc page into structured Markdown for an AI assistant.
How do I install huashu-md-html?
Install the skill with skills.sh: npx skills add alchaincyf/huashu-md-html Install Python dependencies for markitdown: python3 -m pip install 'markitdown[all]' Install pandoc (macOS example from the README): brew install pandoc Convert a markdown file to a styled HTML page: python3 scripts/md_to_html.py article.md --theme article -o article.html
Where can I get huashu-md-html?
The source code and installation instructions are on GitHub: https://github.com/alchaincyf/huashu-md-html

Summary generated by AI from the project's README. GitHub stats refresh daily.

Related projects