ok so I wanted to circle back on this.  I have stitched together a few open source tools to accomplish this, documenting here in case others find it useful.  The majority of my training material is in markdown format.  I've got each markdown file setup to represent a different module of my training course, and the idea was that each of these will become their own scorm package for upload to our LMS.  Doing it this way allows me to have one source format and output to 2 (or more) different formats, which is necessary for me as I definitely see a time when either one of my company's customers or a member of my executive team asks me for a simple, single document that contains all my training material.  If I've got nothing but a bunch of custom scorm packages I can't easily do this, but if the source content is kept in markdown I can do both with _relative_ ease. 

So: the markdown files are converted to html via marked.  Marked is a javascript library that works within a web browser client or can be used via command line to output to stout.  json_encode() is then used to make the html json compliant.  json_encode is a php function.  As mentioned I have 1 markdown file per course, so the bash script, crude as it may be, is as follows:

#!/bin/sh

# quit if input isn't a directory
if [ ! -d "$1" ]
then
        echo 'input not a directory.  Input directory must contain 1 or more markdown files with an .md extension'
        exit 1
fi

for f in $1/*.md
do
        marked $f > $f.html
        json_encode.php $f.html > $f.html.json
done

exit 0
json_encode.php reads as follows:

#!/usr/bin/php
<?php

        $html = file_get_contents($argv[1]);

        print(json_encode($html));

?>

the resulting input.md.html.json file is manually copied into the body tag of the components.json file in the src/course/en folder of my course. 

My content contains images, and I've made this work by ensuring I reference them using relative paths to course/en/images.  non-relative paths will work in the "grunt server" preview using the Adapt framework, but _will not_ in my LMS - only the relative path works. 

So there it is....I hope this is useful to someone.  Scorm is great and all, but if all your content is only in that format...as near as I can tell you would be limited to having your training content in an LMS - you couldn't easily make a simple document out of the same content.  Maybe this is way overkill, maybe not...I'm honestly not sure, but at the very least, perhaps this info will be useful to someone...