I recently shared the release of zaje - a syntax highlighter that aims to cover all your shell colouring needs.
I’m pleased to announce the release of super-zaje which can do everything zaje can but introduces one very useful enhancement: thanks to the magic of tesseract and gosserac, super-zaje is able to extract text right off an image (both HTTP(s) URLs and local files are supported).
The below video demonstrates its usefulness with a few sample LinkedIn posts that shared code as screenshots (a common case, alas, due to LI’s inexplicable refusal to support markdown)
For installation instructions, see README