Clipppy is a convenience layer for inference and probabilistic programming in Python. Originally designed to automate many common tasks related to defining, manipulating, and exploiting forward models (while trying not to be annoying in the meantime), it now incorporates—alongside venerable utilities for variational inference and interfaces to other packages for more conventional Bayesian inference—a somewhat-fledged collection of SBI routines: in fact, it contains the implementations of all primary inference methods presented and used in this thesis, i.e. hierarchical and set-based TMNRE and neural model selection. This chapter briefly introduces Clipppy’s main components, focusing on its SBI-related utilities: model stochastification and prior truncation.