SUPER data type parsing when import data from S3 to Redshift

I have a csv file in which there is 1 column with nested format. Something looks like this:

I want to load this file stored in S3 into Redshift and use "SUPER" as the data type for detail column.

The query i'm using is:

CREATE TABLE test
(id smallint
type varchar(10)
detail super)
diststyle auto

and here is the copy statement (in aws documentation, there's not much mentioned about parsing data to super type):

copy stage.test
from 's3://.../test.csv'
region 'us-east-1' iam_role 'arn:aws:iam::xxxxxxx:role/redshift-role'
ignoreheader 1
format CSV 
;

i could load the data alright, but the SUPER column doesn't seem to be parse correctly. The result in that column looks like a string and i can't unnest it to query the components.

Could someone please tell me where i went wrong?