Originally swapped kefnet out for not being good enough, but would like to re-add with an errata to improve it.

Options for re-balacing: -

Adjusting the cards in hand requirement and/or the draw cost:

• Kefnet the Mindful can’t attack or block unless you have six or more cards in hand. : Draw a card, then you may return a land you control to its owner’s hand.

Changing from "cards in hand" to "unless you've discarded":

• Kefnet the Mindful can’t attack or block unless you've discarded a card this turn.

• : Draw two cards, then you may discard a card. = Big mana sink, but huge rate
• : Draw two cards, then discard a card = worse version of above
• : Draw two cards, then discard a card = better version of above
• : Draw a card, then discard a card
• : Draw a card, then discard a card = looks awesome
  Major downside of this option is that it requires Kefnet's controller to spend mana on their turn if they want to attack.

In the same vein, I feel that Bontu is the other god in the cycle that falls below rate in the current iteration of the cube. Thoughts for bringing his viability up a little bit are:

Swapping to a revolt style trigger:

• Revolt — At the beginning of combat, if a permanent you controlled left the battlefield this turn, Bontu the Glorified can attack and block this turn. , Sacrifice another nonland permanent: Scry 1. Each opponent loses 1 life and you gain 1 life

Revolt itself feels shoehorned in, and having him trigger on fetches is possibly too strong, maybe keeping the text consistent with the other gods is better?:

• Bontu the Glorified can’t attack or block unless a nonland permanent under your control left the battlefield this turn. , Sacrifice another nonland permanent: Scry 1. Each opponent loses 1 life and you gain 1 life.

This change to nonland permanents would allow Bontu to function more reliably in more decks, allowing synergies with an artifact or blink strategy.