Is there a way to quickly process hundreds of data with the kruskal-wallis test?

Reshape your data into a long format and nest by "ko*" variables.

If you need more specific help, please provide a proper REPRoducible EXample (reprex) illustrating your issue.