StellarSand
Add diceware
Some people use diceware as password generation method. The word lists could be included as another dicttionary, so the user could check the strength of the password in case the attacker only tries diceware combinations.
There are many languages that have lists: https://theworld.com/~reinhold/diceware.html#Diceware%20in%20Other%20Languages|outline
Good recommendation, however, the original Diceware lists are full of unnecessary words and junk that would take quite a bit of time to clean up. I just don't have that kind of time right now. For instance, the lists include numbers, repeated characters (e.g., aaa, dd), symbols and many other words, which the app already handles with the library & included dictionaries, so there's no point in duplicating it. Additionally, there are some random, nonsensical words (like "db", "dc", "dt", "em", "gi") that just clutter the list without adding any value. That's why the app already uses the EFF’s passphrase list (which is much cleaner and more focused) for both dictionary matching and passphrase generation. If I ever have the time (& feel adventurous!), I may revisit the Diceware lists, clean them up & include them.
Well, for me the value in diceware is that there are a lot of lists for other languages. Catalan list is pretty decent.
I also noticed another thing when you enter a diceware password, which is it seems that the app seems to overestimate entropy? This is the password generated at https://diceware.rempe.us/#eff. The generator shows 65 bits whereas the app shows 92. So, if use diceware generated passwords and I want to know the number of guesses an attacker would need trying only diceware combinations, the number in the app seems overestimated.
I want to clarify that this app is primarily designed for checking passwords rather than passphrases. The underlying library used, does not support passphrase analysis, which means that results for passphrases will not be accurate. This limitation can lead to discrepancies in entropy calculations (and other results), such as the one you mentioned, where the app shows a higher estimate than the Diceware generator. I’ll keep your suggestion in mind for future updates when I have the opportunity to revisit the implementation.