Scripting the CLI

The examples use bash and jq.

Exit status

megaloader exits 0 on success and 1 when extraction or a download fails. A usage mistake, such as a missing URL, exits 2. It prints a failure message on standard output, not standard error, so test the exit status and not the text. Capture the output first and parse it only when the command succeeded:

if json=$(megaloader extract "$url" --json); then
  echo "$json" | jq .count
else
  echo "failed: $url" >&2
fi

Read the JSON

Number of files:

$ megaloader extract https://bunkr.si/a/xYKtNmBx --json | jq .count
6

Filenames:

$ megaloader extract https://bunkr.si/a/xYKtNmBx --json | jq -r '.items[].filename'
sample-image-06.jpg
sample-image-04.jpg
sample-image-05.jpg
sample-image-03.jpg
sample-image-01.jpg
sample-image-02.jpg

Download URL and Referer as tab-separated columns, for a downloader that you drive yourself:

$ megaloader extract https://bunkr.si/a/xYKtNmBx --json \
    | jq -r '.items[] | [.download_url, (.headers.Referer // "")] | @tsv'
https://c1mp-b.cdn.cr/3e48b142-bc3b-4b8d-bbc9-db9f891687d3.jpg?n=sample-image-06.jpg	https://get.bunkrr.su/
https://c1mp-b.cdn.cr/8c385ba6-9650-4bbe-a4a2-45557ad137e1.jpg?n=sample-image-04.jpg	https://get.bunkrr.su/
...

Total size of the files that report one. A platform that does not report sizes gives 0:

megaloader extract "$url" --json | jq '[.items[].size_bytes // 0] | add'

Download a list of URLs

Put one URL per line in urls.txt, and record the ones that fail:

while read -r url; do
  megaloader download "$url" downloads || echo "$url" >> failed.txt
done < urls.txt

Because download skips files that exist, run the same loop over failed.txt to retry the failures.

Keep a folder up to date

Run the same command on a schedule. Files already on disk are skipped and new files are fetched:

0 6 * * * megaloader download https://gofile.io/d/abc123 /srv/media/abc123 --token "$GOFILE_TOKEN"

Cron does not set your shell environment. Define GOFILE_TOKEN in the crontab or pass the token directly.