can we download each attachment separately using the envelope ID and data label, or is there any other way?

Look into DocuSign Retrieve.  It lets you download the completed envelope and each file inside can optionally be merged into one or kept separate.  It also creates an index file.  When it’s done, you end up with a structure like:

• /Download_Directory
  • index.csv
  • </envelopeid_One>
    • <fileA>
    • <fileB>
    • <certificate_One.pdf>
  • </envelopeid_Two>
    • <fileC>
    • <certificate_Two.pdf>
  • </.downloaded>
    • envelopeid_one.xml
    • envelopeid_two.xml
    • envelopeid_three.xml
    • envelopeid_four.xml

The .downloaded folder is a running tab of which envelope that Rertieve script has looked at.  You can use Retrieve to set your script to run on a schedule.  And your script has a parameter for what timeframe to look at (last X days, for example).  

A typical use is schedule for 1hour between runs, script timeframe parameter is last 72hours.  So your script runs, finds envelopes, then checks their IDs against the .downloaded directory to see if that ID is present.  If it is, the script skips it.  If it’s not, the script checks it against the parameters its looking for and logs the result in .downloaded as an xml file.  In the above example, inside envelope_three.xml, it may say something like isMatch No.  And envelopeid_one would say isMatch yes.  Either way, those two envelopes have been screened by this script once already and don’t need to be screened again next time the script runs.

The 1h/72h setup gives you leeway so that if your server or computer you run the script on happens to go down and come back online later that day, your script will pick up the envelopes completed during that downtime.  3 days/72 hours let’s it stop for the weekend and then start up again on Monday without you having to go manually grab the envelopes.

The script also lets you choose what elements to include in the index.  You can use Custom Fields in eSignature to good effect to populate the index.