Recently, the concept of intent debt has become a big discussion for LLM-related research in software engineering. It basically concerns the "fee" that developers pay after using LLMs too much that at some point no one would understand the intention behind the system, or *why* certain decisions were made. The problem is that we still don't know how intent debt looks like or what fees are to be paid by developers. This project aims to mine software repositories to investigate these questions. The repositories will be later used to create some baselines, and local LLMs will be run on certain tasks and compared against these baselines.