The point 1 is right, the tests are not checking for words converted to camel case, but just to first letter capitalized instead which is highly misleading as it clearly states twice that the words should be converted to camel case

I have redone the wording. If you think it is not good enough please open another issue and be more specific.

2. "-the-stealth-warrior" should be converted to "theStealthWarrior" - You've pointed out problem that was missed in kata's description. I agree that if we should erase dashes/underscores from the front of the text, the lower letter must remain lower.
   From the other hand, do dashes/underscores from the front of the text (back from the text) are should be considered as delimiters at all? They didn't tell us about it too. "-the-stealth-warrior__" -> " "theStealthWarrior" or "-theStealthWarrior__"

Wording of the question is ambiguous.

If you do lowercase the other letters, the random tests will fail because the include random capitals within strings.

There are two problems with this kata:

1. The tests don't check if words are converted to camel case, only that their first letter is capitalized. Some of the programs posted as solutions aren't valid, because they don't lowercase the other letters.
2. According to the description, the "first word within the output" should be capitalized only if the original word was capitalized. In the case that the input begins with a delimiter, then the first word in the output is the word that occurs in the input after this delimiter. The tests are incorrect in this case. Example: according to the description, "-the-stealth-warrior" should be converted to "theStealthWarrior" (where "the" is not capitalized in the output because it's not capitalized in the input). However the tests require "TheStealthWarrior".