The ensembl-io repo is intended as a shared codebase for handling the parsing and writing of popular biological formats used by Ensembl, such as BED, BigWig and FASTA. For a full list of supported formats, see the child objects in modules/Bio/EnsEMBL/IO/Parser/.
As the code matures, it is anticipated that various teams within the Ensembl project will begin to integrate Bio::EnsEMBL::IO modules into their pipelines; it will also be used in future releases of the Ensembl website to handle parsing of uploaded data and output of exported data.
All parsers should have associated unit tests, which can also serve as simple tutorials on how to use ensembl-io.