In addition to the ISO standard (which will be expected in industry and many government contexts), there are the evaluated nuclear data files (of which there are actually several sets as several countries maintain their own), if you are willing to accept data that has already been processed and harmonized by experts.1
The format of the files is a bit painful by modern standard, but if you have processed a lot of data in Fortran 77, you'll be right at home.
1 As opposed to the raw data from individual experiments each with their own systematic and methodological issues. I strongly recommend accepting the processed data. Just once I decided that the data I wanted was obscure and may not have been considered important by the ENDF maintainers and went back to the original sources myself. Waste of three weeks work: my results agreed quite closely with the official results.