In the PPTX Converter convert function, this 'filename' placeholder is being placed inside of the content section for the md image object:

Causing us to not have the ability to support images post process:

In the shape.image object, we have content_type(file type), and blob(bytes). Could easily format a base64 encoded image here instead:

img_b64_string = base64.b64encode(image_bytes).decode('utf-8')
markdown_image = f"data:{content_type};base64,{b64_string}"
md_content += ( "\n![" + (alt_text if alt_text else shape.name) + "](markdown_image)\n" )

This would allow us to process images post conversion. Not sure if it was decided b64 encoded bytes wasn't ideal, but that can easily be extracted/converted/uploaded remotely etc. to a user's needs, and could even pass a flag to make it optionally supported.