This filter counts the words and characters in the body of a document (omitting metadata like titles and abstracts), including words in code. It should be more accurate than wc -w or wc -m run directly on a Markdown document, since wc will also count markup characters, like the # in front of an ATX header, or tags in HTML documents.

To run it, pandoc --lua-filter wordcount.lua myfile.md. The word count will be printed to stdout.

If you want to process the document as well as printing the word count set the variable wordcount to process (or process-anyway or convert). This works only in conjunction with the standalone document option (-s). This can be done through the command line:

pandoc -s -L wordcount.lua -M wordcount=process sample.md -o output.html

Or the document's metadata block:

---
title: My Long Book
wordcount: process-anyway
---