Why Git Never Truly Deletes Files — And What That Means for Secrets

Git stores every commit as a complete snapshot rather than a diff, meaning removed files remain fully recoverable in repository history. Deleting a file and committing the removal does not erase the original blob object, which stays reachable as long as the commit that introduced it exists. Running git gc does not help either, since garbage collection only removes objects that nothing points to. Identical file contents across a repository are automatically deduplicated by Git's content-addressed storage, where each object is named by a hash of its contents. For large files with incremental changes, Git's pack files use delta compression to keep repository size manageable without discarding any historical data.