MArcomage

I just think this card is underrated. Right now it sits at the bottom, despite being a very good card. If you have a deck that that isn't so much focused on summoning rares and uncommons, you should try it ;-)

I could say the same for the old Divine guidance. It was a very good card in a holy deck: +11 resources for some almost useless tokens. AND it was restoration.
Anyway, yes, pixie is a good card I will try to add it in some of my decks.

There are two reasons why this card is underused.
1. This card has very bad hand dependancy. If you have 8 commons - you can't get rid of hand fast enough without Wish and Militia, and you can't do much with this hand. Of course, you can build a rush deck with Pixie - there are many good rushing commons, but there is another drawback...
2. ...this card is double-edged sword. So your enemy will get somewhat equal amount of resources if you'll play it. And in that case - why pick Pixie if you can pick zero-cost Wealth with more stock?
Of course, you can try to cripple Nature or Illusion decks with it, but there are more effective and reliable cards for that. So Pixie is poor choice in any situation.
Oh, and one more thing. You can't really build a deck around Pixie, unlike Moon shrine or Moon castle.

I totally agree with NG. Instead, this card could
1) Only affect you positively,
2) Discard a common card after it's played, or
3) Only affect your opponent negatively either a) based on commons in your hand, or b) based on non-commons in his hand

Just some thoughts.

I agree with NG_Beholder.
Chances are that it will give at least 1-2 stock (maybe more) to the opponent and that's a serious drawback.

This could be better:

Steal 1 random resource for each common card in hand.

I tried using this card for a while in one of my rush decks, but found it unimpressive for basically the reasons that NG gives.

Thanks for feedback. I'll analyze it further when I get home :)

I think I could change it to zero cost and have it stock: +N where N = #common cards in hand / 2 (it also counts self as Working together does). This way it will give 0 to 4 stock. Average number of commons is 3 to 5 most of the time, so that would mean around 2 stock average, but potentially unreliable. Maybe capping it to 3 max stock, or making it N = (#common cards in hand / 2) - 1 or N = (#common cards in hand - 1) / 2

What do you think?

This could be interesting. But max stock shouldn't be capped IMO.

Here are comparisons of multiple variants (firs number is number of common cards in hand, the second one is stock gain):

Variant A: N = #commons in hand / 2

1 - 1
2 - 1
3 - 2
4 - 2
5 - 3
6 - 3
7 - 4
8 - 4

Variant B: N = (#commons in hand / 2) - 1

1 - 0
2 - 0
3 - 1
4 - 1
5 - 2
6 - 2
7 - 3
8 - 3

Variant C: N = (#commons in hand - 1) / 2

1 - 0
2 - 1
3 - 1
4 - 2
5 - 2
6 - 3
7 - 3
8 - 4

I think Variant B is the most balanced.

How about C, but with 2 gems or 2 recruits cost?