Markdown to EPUB

Une compétence d'agent Claude qui convertit des documents Markdown et des résumés de chat en fichiers EPUB professionnels. Idéal pour créer des livres électroniques portables et compatibles avec tous les appareils à partir de documents de recherche, d'articles de blog, d'articles ou de résumés de conversations.

npx skills add https://github.com/smerchek/claude-epub-skill --skill markdown-to-epub-converter

Markdown to EPUB Converter Skill

This skill transforms markdown documents into professional EPUB ebook files. Perfect for converting research documents, blog posts, articles, or chat conversation summaries into portable, device-agnostic ebook formats.

Overview

The skill accepts markdown content in multiple formats and generates a properly formatted EPUB3 file that works across all major ebook readers including:

  • Apple Books
  • Amazon Kindle (via Kindle for Mac/Windows/iOS/Android)
  • Google Play Books
  • Kobo and other EPUB readers
  • Any standard EPUB reader

Input Formats

Option 1: Raw Markdown Text

Provide markdown content directly in your message:

Convert this markdown to EPUB:
# My Book Title
## Chapter 1
This is chapter one content...

Option 2: File Path

Provide a path to a markdown file to be converted.

How It Works

  1. Markdown Parsing: Analyzes your markdown and automatically:

    • Treats H1 headers (#) as chapter boundaries
    • Treats H2 headers (##) as section headings within chapters
    • Preserves formatting (bold, italic, links, lists, code blocks)
  2. Structure Generation: Creates proper EPUB structure:

    • Automatic table of contents from chapters
    • Navigation document (EPUB3 standard)
    • Metadata (title, language, etc.)
  3. File Creation: Generates a valid EPUB3 file ready for download and use

Usage Examples

Example 1: Convert a Blog Post

"Convert this markdown blog post to EPUB:

How to Build a Simple Web Server

Introduction

...content..."

Example 2: Convert a Research Summary

"I have research notes in markdown format. Convert them to an EPUB ebook. The content is:

Research Project: Machine Learning Basics

Chapter 1: Fundamentals

..."

Example 3: Convert a Chat Summary

"Summarize our conversation so far as markdown and convert it to an EPUB for reference"

Output

The skill generates a downloadable EPUB file that includes:

  • Professional formatting
  • Automatic table of contents
  • Proper chapter structure
  • Support for markdown formatting elements:
    • Headers (all levels)
    • Bold and italic text
    • Hyperlinks
    • Lists (ordered and unordered)
    • Code blocks and inline code
    • Blockquotes
    • Horizontal rules

Markdown Elements Supported

ElementMarkdownSupportNotes
Headers# H1 through ###### H6FullAuto TOC generation
Bold**text** or __text__Full
Italic*text* or _text_Full
Links[text](url)FullClickable in ebooks
Lists- item or 1. itemFullNested lists supported
Code blocks```languageEnhancedSyntax highlighting ready, monospace fonts
Inline code code EnhancedStyled background, borders
TablesMarkdown tablesEnhancedStyled headers, alternating rows
Blockquotes> quoteFullStyled with left border
Horizontal rule--- or ***Full

Advanced Features

Enhanced Code Block Support

Code blocks are beautifully formatted with:

  • Premium monospace fonts: SF Mono, Monaco, Fira Code, Consolas, and more
  • Styled backgrounds: Subtle gray background with blue accent border
  • Language detection: Specify language after ``` for future syntax highlighting
  • Proper escaping: HTML characters are safely escaped
  • Overflow handling: Horizontal scrolling for long lines

Example:

def fibonacci(n):
    if n <= 1:
        return n
    return fibonacci(n-1) + fibonacci(n-2)

Enhanced Table Support

Tables are rendered with professional styling:

  • Styled headers: Blue background with white text
  • Alternating rows: Zebra striping for readability
  • Cell padding: Comfortable spacing for easy reading
  • Inline formatting: Code, bold, italic, and links work in cells
  • Responsive: Tables adapt to different screen sizes

Example:

FeatureStatusNotes
HeadersFull support
CodeEnhanced styling
TablesProfessional layout

Custom Title and Metadata

You can specify EPUB metadata:

  • Book title (defaults to first H1 header)
  • Author name
  • Language
  • Publication date

Chapter Organization

Chapters are automatically detected from:

  • H1 headers (#) as primary chapter breaks
  • Logical content sections between H1s
  • Automatic page breaks between chapters

Styling

The generated EPUB uses clean, readable default styling that:

  • Respects the reader's font preferences
  • Works on all screen sizes
  • Maintains proper spacing and hierarchy
  • Includes appropriate margins and padding

Technical Details

  • Format: EPUB3 (compatible with all modern readers)
  • Encoding: UTF-8
  • HTML Version: XHTML 1.1
  • CSS Support: Responsive styling

Downloading Your EPUB

After generation, the file will be available for download. You can then:

  1. Download the EPUB to your computer
  2. Open it with your preferred ebook reader
  3. Transfer to your Kindle, iPad, or other device
  4. Upload directly to Kindle via email or cloud

Tips for Best Results

  1. Use Proper Markdown Structure: The skill works best when markdown follows standard conventions (H1 for titles, H2 for sections)

  2. Clear Chapter Breaks: Use H1 headers to clearly mark chapter divisions

  3. Descriptive Headers: Headers become the table of contents, so make them clear and descriptive

  4. Content Organization: Place content logically between headers

  5. Supported Formatting: Stick to basic markdown formatting for best compatibility across all readers

Troubleshooting

EPUB doesn't open: Ensure your markdown is properly formatted. Check for matching brackets in links and proper syntax.

Table of contents is empty: Make sure your markdown includes H1 headers to define chapters.

Formatting looks different: EPUB readers apply their own fonts and styling. This is normal and expected behavior.

Scripts

  • epub_generator.py - Core EPUB file creation and formatting
  • markdown_processor.py - Markdown parsing and structure extraction

Future Enhancements

  • Auto-generated cover pages with custom images
  • Kindle-specific optimizations (.mobi format)
  • Custom CSS styling per user preferences
  • Multi-document merging
  • Image embedding and optimization
  • Advanced metadata support

Skills connexes

margin-analyzer
anthropic
Analyse l'économie unitaire par produit ou service à l'aide des insights marchands PayPal et des données de coûts QuickBooks, compare avec l'inflation et les variations de coûts, et affiche…
official
legacy-circuit-mockups
github
Maquettes de circuits sur breadboard en HTML5 Canvas pour projets de rétro-informatique et d'électronique vintage. Prend en charge plus de 40 composants, dont les microprocesseurs W65C02S, les temporisateurs 555, les EEPROM 28C256, les VIA W65C22, les portes logiques de la série 7400, et des composants passifs comme les LED, résistances et condensateurs. Affiche les circuits sur une grille de 20 pixels avec rotation des composants, code couleur des fils et tables de référence des brochages pour une disposition précise. Inclut des workflows étape par étape pour les montages courants : circuits à LED, 555...
official
next-compile
vercel
Check Next.js compilation errors via a running dev server. Turbopack only. MANDATORY after every code edit before reporting work complete. Replaces `next…
official
skill-creator
firecrawl
Guide pour créer des compétences efficaces. Cette compétence doit être utilisée lorsque les utilisateurs souhaitent créer une nouvelle compétence (ou mettre à jour une compétence existante) qui étend les capacités de Codex…
official
mapbox-android-patterns
mapbox
Modèles d'intégration officiels pour le SDK Mapbox Maps sur Android. Couvre l'installation, l'ajout de marqueurs, la localisation de l'utilisateur, les données personnalisées, les styles, le contrôle de la caméra, et…
official
figma.audit-connect
coinbase
figma.audit-connect — une compétence installable pour agents IA, publiée par coinbase/cds.
official
gemini-api-cli
google-gemini
Guide pour utiliser l'outil en ligne de commande de l'API Gemini. À utiliser lorsque vous devez interagir avec l'API Gemini via la ligne de commande, gérer des agents ou générer du contenu multimédia (images,…)
official
exploring-llm-traces
posthog
PostHog capture l'activité des agents LLM/IA sous forme de traces. Chaque trace est un arbre d'événements représentant une interaction IA unique — de l'invocation de l'agent de haut niveau jusqu'aux appels individuels à l'API LLM.
official