Converts Word (.docx) documents into Markdown using a bundled Python script that also extracts embedded images into a self-contained folder structure. Only supports .docx files, not legacy .doc.
Tasks
Convert a single .docx file to Markdown by running: python scripts/convert_word_to_md.py "/path/to/document.docx"
Batch convert all .docx files in a folder including subfolders by running: python scripts/convert_word_to_md.py "/path/to/folder" --recursive
Inputs
Path to a .docx file
Path to a folder containing .docx files
Optional -o flag with an explicit output path
Optional --recursive flag to include subfolders in batch mode
Outputs
A per-document folder containing a .md file
An img/ subfolder with extracted images (only if the document has embedded images)
Relative image references in the Markdown formatted as img/imgNNN.ext
Limitations and checks
Only supports .docx files; legacy .doc files are not supported
markitdown embeds images as truncated base64 placeholder URIs rather than real image data
Image placeholders may be left unreplaced if MarkItDown's placeholder count does not match images found in word/media/
The script only produces Markdown and images; it does not interpret content
Verify the source path is a fully resolvable absolute path before running conversion; use ask_user if only a filename or ambiguous path is provided
Confirm that .docx files are being processed and legacy .doc files are rejected with exit code 3
Ensure the output folder is created next to the source file by default and that -o is only passed when the user explicitly provides an output path
When processing a folder containing mixed file types (.pdf, .docx, .xlsx), verify that sibling skills convert-pdf-to-md and convert-excel-to-md are also invoked
AI Search
Find projects, verify facts, compare options, or turn a complex need into an actionable plan
Try a searchA click only fills the search box; you stay in control