All bots that aren't using autoit or aren't kernel drivers (basically every commercial bot outside of $100/month cs:go hacks) inject.
The thing here isn't to be worried that a bot injects, you should be worried that the author says his bot doesn't inject because it probably means he doesn't know how to reverse warden.
There are generally 2 ways to avoid detection.
1) Reverse warden and make sure they aren't targetting what you hook (what buddy bots do).
2) Be obscure and hope that warden doesn't target you specifically or accidentally target what you hook (what smartbot does).
However, smartbot has never had a ban wave while hearthbuddy has. The staff here claims it wasn't software detected, it might have been due to the fact that hearthbuddy moved really slowly (just today HB attack speed was increased) and acted very bot-like a year ago. It is better now, but who knows what the real cause was?
It would be nice if the HB ai was smarter, but it seems like they don't want to improve it much since its open-source and other bots could copy it. And they don't want to buy servers to make calculations server-side which might be expensive for them, but it would shut down c-r-a-c-k-ed copies.
Last edited: