This is achieved by analyzing your Git commit history, and finding hot-spots in your codebase.

Use case ideas: - improve "live-grep" experience by prioritizing "trending" results (this is my use case, I use it in telescope.nvim) - find certain patterns, such as TODO comments, specifically in stale code - attempt to feed the most relevant results to an LLM or AI agent when the context window does not allow fitting your entire codebase into the context window - before doing find-and-replace across files, look at the "least expected" examples to see how what would be replaced in areas of the codebase people in the team rarely think about