When importing the CSV, Jira/Xray processes up to 55 rows without any issues. As soon as the file contains 56 rows, the importer misclassifies one of the step rows as a separate issue. Since step rows do not contain a Summary field, this leads to summary='null' and the import fails. The test cases themselves are created correctly, but the importer aborts during step creation once the total number of rows exceeds this limit.