EPUB to Markdown

This tool reads an EPUB and rewrites its reading-order text as a single Markdown (.md) file. It is built for moving a book's text out of a closed e-reader format and into tools that speak Markdown: importing a title into Obsidian or Notion, dropping chapters into a static-site generator (Hugo, Jekyll, Astro) as content source, or keeping a plain, portable text copy of something you own.

It suits writers and researchers who want to quote, annotate, or restructure a book's prose in a Markdown editor, and anyone archiving an e-book in a format that stays readable without special software. Because chapter titles become real Markdown headings, the output keeps a navigable structure rather than collapsing into one undifferentiated block of text.

It is not a layout tool: to keep the book's look, keep the EPUB; Markdown is for reusing the text.

Drop or select an .epub file. Everything happens in your browser: the file is unzipped and parsed locally, walked in reading order, and each chapter's HTML is mapped to Markdown. Nothing is uploaded, and the parsed book is gone once you close the tab.

The converter maps structure where it can. Headings h1 through h6 become # to ######; bold (strong or b) becomes **text** and italic (em or i) becomes *text*; lists, blockquotes and inline code are kept, and code blocks are fenced with their spacing untouched; tables become pipe tables of their cell text, with the first row as the header; links to web addresses keep their URL, while links between the book's own pages become plain text. Each chapter starts with its title as a heading, from the table of contents or the first heading. You get a chapter checklist with All/None toggles and word counts, optional YAML front matter (title, author, language, date), then Download .md or Copy to clipboard; chapters are joined with --- lines.

What it does not do: it does not export images. An image with alt text becomes an [Image: …] note so you can see where it was; one without alt text leaves nothing. DRM-protected EPUBs are detected and refused.

Fidelity depends on how the EPUB was authored. Books with clean semantic HTML (real headings, strong and em tags) convert cleanly; books that fake formatting with CSS classes lose that emphasis, since the tool reads tags, not stylesheets. Nested lists are flattened into their parent item, table cells keep their text but not inline formatting or merged-cell layout, and footnotes stay wherever they sit in the text.

A common surprise is chapter naming. If the book's table of contents is thin, some sections take their first heading, page title or file name as a title, and front matter or copyright pages appear as their own chapters. Untick anything you do not want before exporting rather than deleting it afterward.

Respect copyright too: extracting text from a book you may not redistribute is on you, not the converter.

Open the file in your Markdown editor and skim top to bottom. Confirm chapters are in the right order and that headings render at sensible levels (a chapter title showing as ### instead of # usually means the source used a lower heading). Spot-check a few paragraphs for dropped emphasis, and look at the start and end for copyright or other pages you meant to exclude. If you enabled front matter, verify the title and author fields populated.

Search for [Image: to find where pictures were, and decide whether to add them back. Check that list items, tables and blockquotes kept their shape, and that the --- chapter separators land between chapters rather than mid-text.

Before quoting or republishing, compare key passages with the original EPUB, since styled text and footnotes can shift.