Doc to Markdown Skill

doc-to-markdown is the inverse of pdf-creator and docx: when you already have a document and want it in plain Markdown — for editing, republishing, archiving, or feeding into another AI workflow.

Supported inputs

• .doc — legacy Word format
• .docx — modern Word
• .pdf — text-based or scanned (with OCR fallback)
• .pptx — slide content extracted as Markdown sections

What’s preserved

• Heading hierarchy — H1 / H2 / H3 from styled headings
• Lists — bulleted and numbered
• Tables — as Markdown tables
• Embedded images — extracted to a ./figures/ folder, referenced by relative path
• Footnotes / endnotes — preserved as markers

Where it shines

• Migrating content from old Word / PDF docs into a Markdown-based system
• Republishing archived material on new platforms
• Feeding AI workflows that prefer Markdown input
• Preparing for translation — Markdown round-trips translation tools cleanly

Composes with

• pdf-creator — convert Markdown back to PDF if needed
• wechat-article-writer — clean source material for a new article
• humanizer-zh — strip AI-detection signatures from extracted text

→ Install on GitHub

Use doc-to-markdown to convert this 60-page consultancy report (.docx)
into Markdown. Save embedded images to ./figures/. Preserve the heading
hierarchy and footnote markers.