Okay, you used some good balance theories, like +1 cost for each attack, and another raise for more-than-average HP. You also made an estimate on the ability. But will it work?
Things to consider:
The cost/size of the creature. If the size is very large for a petty ability, the creature may feel too 'bulky'. For example, it is relatively easier to use Quintessence on the key creatures than to use Anubis in most cases.
The costs ratio. Remember, the cheaper the creature, the faster the ability may be available. Increasing the cost of the ability may help lower the cost of the creature so that you can play it relatively sooner. For example, a 2-2 cost can be faster than a 3-1 cost, because it leaves the flexibility of having only 2 pillars to start, however the latter allows ease of reuse. Faster tends to be best for one-use abilities, while lesser abilities that can be very effective after multiple uses are best when they are cheap. This applies a lot more heavily when the ability is of a duo cost.
Think it needs more oomph like a mid-range attacker? Messing around with ability costs and drawbacks are a great way to reduce the cost of the creature so that it has a better feel in those vanilla situations. Sometimes, you can achieve what I coin as pseudo vanilla, since it can be built around for pure damage, while also having an ability for the right situations, or to simply augment the already big punch, like Lava Golem.
