BEC/Fraud: Fake investment outreach from suspicious TLD

Detects unsolicited investment/funding outreach emails from suspicious TLDs. Targets mass-mailed spam campaigns offering business funding, capital allocation, and family office outreach.

Sublime rule (View on GitHub)

 1name: "BEC/Fraud: Fake investment outreach from suspicious TLD"
 2description: |
 3  Detects unsolicited investment/funding outreach emails from suspicious TLDs.
 4  Targets mass-mailed spam campaigns offering business funding, capital allocation,
 5  and family office outreach.  
 6type: "rule"
 7severity: "medium"
 8source: |
 9  type.inbound
10  // subject contains investment language
11  and (
12    strings.istarts_with(subject.base,
13                         'financing for',
14                         'investment into',
15                         'an investment opportunity',
16                         'investing in'
17    )
18    or any([subject.base, body.current_thread.text],
19           strings.icontains(., "family office")
20    )
21  )
22  // financial body content
23  and (
24    any(ml.nlu_classifier(body.current_thread.text).topics,
25        .name == "Financial Communications" and .confidence != "low"
26    )
27    or (
28      any(ml.nlu_classifier(body.current_thread.text).topics,
29          .name == "B2B Cold Outreach" and .confidence != "low"
30      )
31      and regex.icontains(body.current_thread.text,
32                          'funding',
33                          'capital',
34                          '\d{3}k',
35                          'rates from \d+%',
36                          'funded in \d+',
37                          'family office'
38      )
39    )
40  )
41  // suspicious sender
42  and (
43    sender.email.domain.tld in $suspicious_tlds
44    or sender.email.domain.tld in ("info", "de")
45  )  
46tags:
47 - "Attack surface reduction"
48attack_types:
49  - "BEC/Fraud"
50tactics_and_techniques:
51  - "Social engineering"
52detection_methods:
53  - "Header analysis"
54  - "Sender analysis"
55  - "Content analysis"
56  - "Natural Language Understanding"
57id: "5d4c4a15-661c-5fd1-9d5f-1c72c8230be8"

Related rules

to-top