My lab practice is to publish the repo path and the full hash of the commit that gave the results in the paper. The repo might have subsequent bug fixes, new co

My lab practice is to publish the repo path and the full hash of the commit that gave the results in the paper. The repo might have subsequent bug fixes, new code or data, etc. But you can get and run the exact code & data that the claims are based on if you wish. The published paper contains the hash printed in the text.