The first thing you need to define is what you mean by 'equivalent'
If you mean 'what you can do with one, you can do with the other' then one approach would be to prove that both languages are Turing complete If you can do this then you will have shown that both languages can perform the same kinds of computations as each other (and any other Turing complete language or device)
If you mean 'they have the same structure - just different ways of specifying elements' then you will need to abstract the languages to show that they do indeed share the same structure. Backus Naur Form is one way of doing this. If two languages have the same structure in BNF with just the terminals being different then really they are just two different representations of the same abstract syntax.
There are obviously other possible meaning of 'equivalent' and thus other things you could do.
You also need to define what you mean by 'proving' - do you mean a rigorous mathematical proof or 'convince my co workers that the scripting language is good enough'? If you mean the first, it's going to be tough. If you mean the second, you could define a specification for what thse languages need to be able to do and demonstrate proof of concepts in each language to show that they can meet the specification.