Using kraken2 to extract/group the read/contigs based on their corresponding taxa (some kind of binning).

[kaydadaemon]

I'm currently using kraken2 to profile metagenomic data, like everyone else.
However, I wanted to use kraken2 to first assign the taxa of contigs for the assembled metagenome and extract/group the contigs based on their corresponding taxa.
A crude method I thought of was to create numerous databases for individual taxa and use the --classified-out flag. The problem with this method I think results in the same contigs hitting multiple taxa.
I wanted to know if there was a way to achieve this in kraken2.