x2md-skill

name	x2md-skill
description	Batch convert a directory of slides and PDFs into Markdown. Supports ppt/pptx/pps/ppsx/odp to pdf+md, and text-based pdf to md (no OCR).

X2MD Skill

Use this skill when the user wants batch conversion to Markdown from common document formats.

What this skill does

• Converts slide files to PDF with LibreOffice (soffice --headless)
• Extracts slide text and notes to Markdown from PPTX
• Extracts text from PDF to Markdown (text PDFs only, no OCR)
• Handles mixed extensions in one directory: .ppt, .pptx, .pps, .ppsx, .odp, .pdf
• Mirrors source directory structure in output

Required toolchain

• Python 3.9+
• python-pptx (pip install python-pptx)
• pypdf (pip install pypdf)
• LibreOffice CLI (soffice in PATH)

Run

python3 /home/ares/yysnotes/x2md-skill/scripts/convert_x2md.py \
  /path/to/input \
  --output-dir /path/to/out \
  --recursive

Output layout

• OUTPUT/pdf/**/*.pdf
• OUTPUT/md/**/*.md

Notes

• For legacy .ppt, the script first converts to .pptx (temp file) so Markdown can be extracted.
• If a slide deck cannot be parsed to Markdown, PDF conversion still proceeds.
• For .pdf, this skill only extracts embedded text. Scanned/image PDFs are skipped (OCR not included).