verdicts contains a list of maliciousness classifications for the file under study based on its behaviour. It is a list of strings, that can contain any of the following:

CLEAN clean, whitelisted or undetected.
MALWARE should be detected as malware.
GREYWARE PUA, PUP (possibly unwanted program).
RANSOM Ransom or crypter.
PHISHING Tries to phish the user or deceive him to steel his credentials.
BANKER banking trojan malware.
ADWARE displays unwanted advertisements.
EXPLOIT contains or runs an exploit.
EVADER contains logic to evade analysis.
RAT remote access trojan, may listen for inbound connections.
TROJAN trojan or bot.
SPREADER spreads to USB, other drives, network, etc. Work-like functionality.

{
    "data": {
        "attributes": {
            "verdicts": [
                "<string>"
            ]
        }
    }
}

{
    "data": {
        "attributes": {
            "verdicts": [
                "MALWARE"
            ]
        }
    }
}