XL-MS output ought to include two scan number columns to support MS3 identification?

[hbarsnes]

The XL-MS output files, e.g. XL_Interlinks.tsv, only contain one Scan Number column (in addition to the Precursor Scan Number column). This is fine if both of the crosslinked peptides were identified in the same MS2 spectrum. However, what if the peptide identifications instead happen in the MS3 spectra? In that case, the alpha peptide will have one (MS3) scan number and the beta peptide a different (MS3) scan number.

This does not seem to be supported in the current tsv-based output formats? Hence, there is currently no way of knowing whether the identification of the two peptides was based on the single MS2 spectrum or on the two MS3 spectra? Or is there something I'm missing?