I'm looking at the v1.5 document.

Things you might like:

• It's a system-neutral way to give a qualitative description of a character - personality, traits, etc.
• It's easy to mix and match different elements, rolling up some and assigning others non-randomly.
• The brief personality descriptions give you a good thumbnail picture of each type: desires, temptations, fears, etc. For some characters, the personality description alone may be sufficient, and you won't need to roll up the other elements.
• Similarly, the Virtue and Vice descriptions are brief but also flexible and useful. Rolling up either or both might sufficient for some characters, without having to roll up the other elements.
• The Stat Archetypes are descriptive and system-neutral, not numeric, but it'll probably be easy to figure out which character stats you should nudge up or down.

Things you might not like:

• If you're looking for a quantitative character generator (rolling up specific stats and selecting a particular species), you'll want another tool. That's not what this is.
• Since it's system-neutral and setting-neutral, there's nothing to factor in differences between species (which ones are taller, stronger, faster, ...). Obviously, you could decide for yourself that Hobbits will tend toward certain personalities and Wookiees toward others, but that's on you. This tool makes no distinction.
• The probability distributions are odd here and there. In the d10, d12, and d20 columns, the tables make the middle rows more likely than the upper or lower rows. So far so good, if the intent was indeed to make the middle rows more likely. The 3d6 column, however, makes the top row the most likely outcome. That is, rolling 3-7 on 3d6 is more likely than any other table result (probabililty 36 out of 216 instead of 27 or lower for other results); that seems like a mistake, or at least it's an unexplained inconsistency with the other tables. Another unexplained probability quirk is that the d100 column makes the middle row the least likely result instead of the most likely, with only a 3% chance instead of the 11-13% chance for each of the other rows. These unexplained quirks seem more like mistakes than design features, but if they're deliberate, it would help if the text pointed them out.
• The Encounter column (which is called Common Professions in the text) is fairly brief and generic. If you have a setting, and if you want a table of common professions, you'll probably want to cook up your own list. Also, note that because of the probability quirks mentioned above, farmers are either the most likely or the least likely profession, depending on which dice you use. Or guards can be anywhere from least likely to most likely, again depending on the dice you use.
• Mistake: The Vices table lists Crazy as the bottom entry. The text explanation uses Deceitful instead. The obvious fix is to decide on your own whether the character is Crazy or Deceitful. I mention this because it looks like a mistake the author might want to fix in a future version.
• Mistake: Every d12 column leaves out a roll of 4. This too has an easy DIY fix, but I mention it in case the author wants to fix the tables.