I guess the dhindo and fapar are mostly flour from wikipedia https://en.wikipedia.org/wiki/Dhindo

Which we do know the nutritional value for, so that part might not be hard to figure out, i.e. write a few recipes, figure out the ingredients and quantities and you can calculate calorific value easily.

Taking a step back, what problem does this solve for your users?

Can we find evidence that people need this, and is there some outcome you want for your users that anything you build would actually help them achieve.

Are Nepalese particularly health conscious, and armed with mobile phones that can download apps from whatever app store you are targeting?

I.e. try to think about whatever your build, even if you gave your users the perfect version of whatever you built, would that change their behaviour or add value to them by helping them do something easier/cheaper/faster or make their life better (for healthcare stuff like calorie counters).