Compare business logic across analyzed commits
See how rules, conditions and product behavior changed — whether the code came from people or AI.
Every time GitMir reads a repository it records what the product does at that moment. Put two of those readings side by side and you get what the product started doing, stopped doing, renamed, or now does differently — in the product's own words rather than as a file diff. A push that only moved code around says so in words, instead of showing you an empty screen.
Open a repository GitMir has read and look at what moved since your last visit.
Pick any two readings to compare, or take the most recent pair.
Read the four lists: what appeared, what went away, what was renamed, what behaves differently.
Ask your agent the same question over MCP when you want the answer while you work.
It compares readings of the repository, not individual commits: if five commits land between two readings, they are compared as one step. What is kept is what each part of the product is called and how it behaves — no lines of code and no file names, deliberately. The record starts at the first reading, so nothing from before you connected the repository is in it, and it keeps the most recent 200 readings.
See it on a real product.
Three models are published in full and open without an account. Ask one of them the question this page is about.