← Back to Explore
sublimemediumRule
Link: Mismatched free file host links with document lure
Detects inbound emails containing mismatched hyperlinks where the displayed URL references a free file hosting service but the actual destination points to another free file host or a suspicious TLD. The rule combines this with NLU classification identifying BEC or credential theft intent, along with body text patterns common to fake document/scan notifications (e.g., 'scanned from', 'shared via', 'for your review') or short, urgency-driven messages. Trusted senders with passing DMARC are excluded to reduce false positives.
Detection Query
type.inbound
and any(filter(body.current_thread.links, .mismatched),
any([.display_url.domain.domain, .display_url.domain.root_domain],
. in $free_file_hosts
)
and (
any([.href_url.domain.domain, .href_url.domain.root_domain],
. in $free_file_hosts
)
or .href_url.domain.tld in $suspicious_tlds
)
)
and any(ml.nlu_classifier(body.current_thread.text).intents,
.name in ("bec", "cred_theft") and .confidence != "low"
)
and 2 of (
regex.icontains(body.current_thread.text,
'scanned from',
'total images',
'attachment format',
'scan information',
'statement/remittance',
'versalink',
'made a file available',
'document from .{0,40}is available',
'(?:uploaded|shared) via',
'(?:report|available) for your review'
),
regex.icontains(body.current_thread.text, '={5,}|_{10,}'),
length(body.current_thread.text) < 700,
regex.icontains(body.current_thread.text,
'kindly review',
'review the attached',
'let us know the next step',
'for your review',
'please review the'
)
)
and not (
sender.email.domain.root_domain in $high_trust_sender_root_domains
and coalesce(headers.auth_summary.dmarc.pass, false)
)
Data Sources
Email MessagesEmail HeadersEmail Attachments
Platforms
email
Raw Content
name: "Link: Mismatched free file host links with document lure"
description: "Detects inbound emails containing mismatched hyperlinks where the displayed URL references a free file hosting service but the actual destination points to another free file host or a suspicious TLD. The rule combines this with NLU classification identifying BEC or credential theft intent, along with body text patterns common to fake document/scan notifications (e.g., 'scanned from', 'shared via', 'for your review') or short, urgency-driven messages. Trusted senders with passing DMARC are excluded to reduce false positives."
type: "rule"
severity: "medium"
source: |
type.inbound
and any(filter(body.current_thread.links, .mismatched),
any([.display_url.domain.domain, .display_url.domain.root_domain],
. in $free_file_hosts
)
and (
any([.href_url.domain.domain, .href_url.domain.root_domain],
. in $free_file_hosts
)
or .href_url.domain.tld in $suspicious_tlds
)
)
and any(ml.nlu_classifier(body.current_thread.text).intents,
.name in ("bec", "cred_theft") and .confidence != "low"
)
and 2 of (
regex.icontains(body.current_thread.text,
'scanned from',
'total images',
'attachment format',
'scan information',
'statement/remittance',
'versalink',
'made a file available',
'document from .{0,40}is available',
'(?:uploaded|shared) via',
'(?:report|available) for your review'
),
regex.icontains(body.current_thread.text, '={5,}|_{10,}'),
length(body.current_thread.text) < 700,
regex.icontains(body.current_thread.text,
'kindly review',
'review the attached',
'let us know the next step',
'for your review',
'please review the'
)
)
and not (
sender.email.domain.root_domain in $high_trust_sender_root_domains
and coalesce(headers.auth_summary.dmarc.pass, false)
)
attack_types:
- "BEC/Fraud"
- "Credential Phishing"
tactics_and_techniques:
- "Free file host"
- "Social engineering"
detection_methods:
- "Natural Language Understanding"
- "URL analysis"
- "Content analysis"
id: "19836dcf-cde4-5d9a-903f-c74c1e88e33d"