Are successes determined with or without cost discounting?

[pstahlhofen]

In RL literature, discount factors are often viewed as part of the definition of standard or constrained Markov Decision Processes. In the standard implementation, however, discount factors are often treated as hyperparameters of RL algorithms. Hence, I am wondering if safety gymnasium computes the sum of discounted costs throughout an episode to determine success or if this sum is taken without any discounting. In the former case, I'd like to know which discount factor is used internally and if it can be customized. I searched both the documentation and the source code for success and was not able to find a clear answer so far