Looking at the code that is currently in the GOLD NMDC data transformation pipeline here: https://github.com/microbiomedata/sample-annotator/blob/main/sample_annotator/clients/gold_nmdc_pipeline.py we can see that the mapping between the GOLD fields and the NMDC slots is mostly hardcoded, with mapping being specified in code.
This issue is requesting for the addition of a declarative mapping file, be it a SSSOM file, a simple TSV file, etc. that specifies the GOLD fields in the first column and the NMDC Schema slots in the second column. The mapping between the fields in the above pipeline should be picked up from the mapping file.