sublimelowRule

BEC/Fraud: Generic scam attempt to undisclosed recipients

Detects potential generic scams by analyzing text within the email body and other suspicious signals.

MITRE ATT&CK

initial-access

Detection Query

type.inbound

// undisclosed recipients
and any([recipients.to, recipients.bcc, recipients.cc],
        any(., .display_name =~ "Undisclosed recipients")
)

// mismatched sender (from) and Reply-to
and any(headers.reply_to,
        length(headers.reply_to) > 0
        and all(headers.reply_to,
                .email.domain.root_domain != sender.email.domain.root_domain
        )
)

// generic recipient
and any(ml.nlu_classifier(body.current_thread.text).entities,
        .name == "recipient" and regex.icontains(.text, "(sir|madam)")
)

// request made
and any(ml.nlu_classifier(body.current_thread.text).entities,
        .name == "request"
)

// not a bec scam
and all(ml.nlu_classifier(body.current_thread.text).intents, .name != "bec")

// negate highly trusted sender domains unless they fail DMARC authentication
and (
  (
    sender.email.domain.root_domain in $high_trust_sender_root_domains
    and not headers.auth_summary.dmarc.pass
  )
  or sender.email.domain.root_domain not in $high_trust_sender_root_domains
)
and (
  (
    profile.by_sender().prevalence in ("new", "outlier")
    and not profile.by_sender().solicited
  )
  or (
    profile.by_sender().any_messages_malicious_or_spam
    and not profile.by_sender().any_messages_benign
  )
)
and not profile.by_sender().any_messages_benign

Data Sources

Email MessagesEmail HeadersEmail Attachments

Platforms

Raw Content

name: "BEC/Fraud: Generic scam attempt to undisclosed recipients"
description: |
  Detects potential generic scams by analyzing text within the email body and other suspicious signals.
type: "rule"
severity: "low"
source: |
  type.inbound
  
  // undisclosed recipients
  and any([recipients.to, recipients.bcc, recipients.cc],
          any(., .display_name =~ "Undisclosed recipients")
  )
  
  // mismatched sender (from) and Reply-to
  and any(headers.reply_to,
          length(headers.reply_to) > 0
          and all(headers.reply_to,
                  .email.domain.root_domain != sender.email.domain.root_domain
          )
  )
  
  // generic recipient
  and any(ml.nlu_classifier(body.current_thread.text).entities,
          .name == "recipient" and regex.icontains(.text, "(sir|madam)")
  )
  
  // request made
  and any(ml.nlu_classifier(body.current_thread.text).entities,
          .name == "request"
  )
  
  // not a bec scam
  and all(ml.nlu_classifier(body.current_thread.text).intents, .name != "bec")
  
  // negate highly trusted sender domains unless they fail DMARC authentication
  and (
    (
      sender.email.domain.root_domain in $high_trust_sender_root_domains
      and not headers.auth_summary.dmarc.pass
    )
    or sender.email.domain.root_domain not in $high_trust_sender_root_domains
  )
  and (
    (
      profile.by_sender().prevalence in ("new", "outlier")
      and not profile.by_sender().solicited
    )
    or (
      profile.by_sender().any_messages_malicious_or_spam
      and not profile.by_sender().any_messages_benign
    )
  )
  and not profile.by_sender().any_messages_benign
attack_types:
  - "BEC/Fraud"
tactics_and_techniques:
  - "Social engineering"
detection_methods:
  - "Content analysis"
  - "Header analysis"
  - "Natural Language Understanding"
  - "Sender analysis"
id: "5dac401f-d3c1-5092-aba4-58ab597c702b"