How to add a field to a csv after deriving its value in a read line

I'm attempting to process in a bash script using awk/cut/sed a simple CSV that looks like this

id,version
84544,abcd v2.1.0-something
3439,abcd a82f1a
3,abcd 2.2.1-bar

Where

1. abcd is constant
2. the string after the space may include a single letter "v" prefix
3. the string after the space may have a build suffix starting with a dash, and
4. the string may instead be a 6 character hex string.

I'm trying to add a field to an output CSV breaking down the second field into only the stripped version number or the hex string.

I've figured out how to parse each line and strip the data I'd like using this

while read line; do
    awk -F '[, ]' '{print $3}' | sed 's/^v//' | cut -d- -f-1;
done < input.csv > output.csv

I'm having trouble figuring out how to construct the output into a CSV like this

id,version,simple_version
84544,abcd v2.1.0-something,2.1.0
3439,abcd a82f1a,a82f1a
3,abcd 2.2.1-bar,2.2.1

Any guidance on constructing the output?

Assumptions:

• commas only show up as delimiters (ie, commas are not part of the data)
• the 1st field never contains spaces
• the 2nd field never starts with spaces
• the 2nd field only has the single space between 2 strings (as shown in OP's sample data)
• existence of the literal string abcd has no bearing on which/how rows are processed (otherwise OP will need to update the question with additional details and sample data)

NOTES:

• once you bring awk into the mix there's usually no need for sed and cut
• since awk is designed for line-by-line processing there's typically no need for a (bash) while loop

One awk idea that replaces all of OP's current code:

awk '-F[, ]' '
{ split($3,a,"-")                                  # split 3rd field on "-" delimiter
  sub(/^v/,"",a[1])                                # strip a leading "v" from first split field
  print $0 "," (NR==1 ? "simple_version" : a[1])   # conditionally print the 3rd field
}
' input.csv > output.csv

#######
# or as a one-liner:

awk '-F[, ]' '{split($3,a,"-"); sub(/^v/,"",a[1]);print $0 "," (NR==1 ? "simple_version" : a[1])}' input.csv > output.csv

An alternative that eliminates the split() by expanding the FS definition:

awk '-F[, -]' '
{ f3=$3
  sub(/^v/,"",f3)
  print $0 "," (NR==1 ? "simple_version" : f3)
}
' input.csv > output.csv

#######
# or as a one-liner:

awk '-F[, -]' '{f3=$3; sub(/^v/,"",f3); print $0 "," (NR==1 ? "simple_version" : f3)}' input.csv > output.csv

Both of these generate:

$ cat output.csv
id,version,simple_version
84544,abcd v2.1.0-something,2.1.0
3439,abcd a82f1a,a82f1a
3,abcd 2.2.1-bar,2.2.1

NOTES:

• keep in mind this is very fragile and will break if any of the assumptions are wrong
• if any of the assumptions are wrong then OP will need to provide a more representative set of their real world data