A CLI utility that converts MHT/MHTML files into clean Markdown, extracting embedded images and resources to a local assets folder with relative links.

• Python >= 3.12

git clone https://github.com/kgoldberg/md-convert.git
cd md-convert
python -m venv .venv
source .venv/bin/activate
pip install -e .

For development (includes pytest):

pip install -e ".[dev]"

md-convert INPUT.mht

This parses the MHT file, extracts embedded resources to an assets/ directory, and writes the Markdown output to page.md (same name as input with .md extension).

Convert an MHT file (output written to page.md):

md-convert page.mht

Convert to a specific output file:

md-convert page.mht -o notes/page.md

Convert with a custom assets folder:

md-convert page.mht -o page.md -a images

1. Parse -- Reads the MHT/MHTML file using Python's email module and validates the multipart/related MIME structure. Handles non-standard charsets like unicode (UTF-16) from Microsoft Word/Outlook.
2. Extract -- Writes embedded images and resources to the assets folder, deduplicating files and handling filename collisions.
3. Rewrite -- Updates <img> and <link> references in the HTML to point to the extracted local assets.
4. Clean -- Pre-processes Word/Outlook HTML by converting CSS-class headings to proper HTML headings, unwrapping layout tables, and stripping MSO conditional comments.
5. Convert -- Transforms the cleaned HTML into Markdown using markdownify with ATX-style headings and dash-style bullets.

source .venv/bin/activate
pip install -e ".[dev]"
pytest

See LICENSE for details.