Spam: Cryptocurrency airdrop/giveaway
Detects messages promoting cryptocurrency airdrops, token claims, or wallet-related rewards.
Sublime rule (View on GitHub)
1name: "Spam: Cryptocurrency airdrop/giveaway"
2description: "Detects messages promoting cryptocurrency airdrops, token claims, or wallet-related rewards."
3type: "rule"
4severity: "low"
5source: |
6 type.inbound
7 and any(ml.nlu_classifier(body.current_thread.text).topics,
8 .name in ("Financial Communications")
9 )
10 and not any(ml.nlu_classifier(body.current_thread.text).topics,
11 .name in (
12 'Newsletters and Digests',
13 'News and Current Events',
14 'Legal and Compliance'
15 )
16 )
17 and not any(ml.nlu_classifier(body.current_thread.text).intents,
18 .name == "benign" and .confidence == "high"
19 )
20 // action word
21 and (
22 strings.ilike(subject.base, '*airdrop*')
23 or strings.like(body.current_thread.text, '*airdrop*')
24 or strings.ilike(subject.base, '*giveaway*')
25 or strings.like(body.current_thread.text, '*giveaway*')
26 or regex.icontains(body.current_thread.text, 'receive \$\d{2,5} worth')
27 or regex.icontains(body.current_thread.text,
28 'claim your (allocation|airdrop|bonus|share|\$\d{2,5})'
29 )
30 or regex.icontains(body.current_thread.text, 'connect .{0,20} wallet')
31 )
32 and not regex.icontains(body.current_thread.text, '\$\d{2,4} (off|cash)')
33 and (
34 // crypto keyword
35 regex.icontains(body.current_thread.text,
36 '\bmetamask\b',
37 '\bethereum\b',
38 '\bbinance\b',
39 '\bgemini\b',
40 '\bwallet\b',
41 '\bkraken\b',
42 '\bsolana\b',
43 '\btrezor\b',
44 '\bledger\b'
45 )
46 // token name, e.g. $USDT
47 or regex.contains(body.current_thread.text, '\s\$[A-Z]{3,4}\s')
48 )
49 and not (
50 sender.email.domain.root_domain in (
51 "gemini.com",
52 "ledger.com",
53 "binance.com",
54 "trezor.io",
55 "kraken.com",
56 "solana.com",
57 "metamask.com",
58 "ethereum.org",
59 "bloomberg.com"
60 )
61 and headers.auth_summary.dmarc.pass
62 )
63attack_types:
64 - "Spam"
65tactics_and_techniques:
66 - "Social engineering"
67 - "Impersonation: Brand"
68detection_methods:
69 - "Content analysis"
70id: "80a2e2fd-6ba1-5989-b9c5-d1e515f3dc82"