Microsoft 365 inbox rule change
Any Microsoft 365 mailbox inbox rule was created, modified, removed, enabled, or disabled (New-InboxRule / Set-InboxRule / UpdateInboxRules /…
What it detects
Any Microsoft 365 mailbox inbox rule was created, modified, removed, enabled, or disabled (New-InboxRule / Set-InboxRule / UpdateInboxRules / Remove-InboxRule / Enable-InboxRule / Disable-InboxRule). Inbox-rule tampering is the signature post-BEC move: a compromised mailbox gets a rule that forwards or hides mail, and the attacker later removes or disables it to cover their tracks. This rule surfaces the WHOLE lifecycle as a lead — including plain creations and edits that carry no forwarding/deletion keywords, which the high-severity forwarding (m365_inbox_forwarding_rule) and hiding (m365_mailbox_hiding_rule) rules do not fire on. Deliberately medium/low-confidence because benign rule changes are common; the escalation is an inbox-rule change from a client IP the offline IOC / reputation sweep has flagged, or from an unfamiliar IP right after a suspicious sign-in — correlate this alert's source_ip against those findings.
MITRE ATT&CK mapping
False positives and triage
Users and admins create, edit, and delete their own inbox rules constantly, so this fires broadly and is a LEAD, not a verdict — expect to tune it per tenant. The high-fidelity signal is an inbox-rule change whose source_ip matches the case's IP/IOC blocklist sweep or a suspicious sign-in for the same mailbox, or a forwarding/deletion rule (already raised at high severity by the dedicated rules). Confirm the mailbox owner made the change and read the rule name/actions in the event evidence before escalating.
The rule
Declarative YAML read by the detection engine — data, never executed. Confidence 40%. Source: rules/cloud/m365_inbox_rule_change.yaml, licensed under the Apache License 2.0.
id: cloud.identity.m365_inbox_rule_change
version: 1.0.0
status: stable
severity: medium
confidence: 0.4
platforms: [application]
mitre_techniques: [T1564.008, T1114.003]
description: |
Any Microsoft 365 mailbox inbox rule was created, modified, removed, enabled, or
disabled (New-InboxRule / Set-InboxRule / UpdateInboxRules / Remove-InboxRule /
Enable-InboxRule / Disable-InboxRule). Inbox-rule tampering is the signature
post-BEC move: a compromised mailbox gets a rule that forwards or hides mail, and
the attacker later removes or disables it to cover their tracks. This rule
surfaces the WHOLE lifecycle as a lead — including plain creations and edits that
carry no forwarding/deletion keywords, which the high-severity forwarding
(`m365_inbox_forwarding_rule`) and hiding (`m365_mailbox_hiding_rule`) rules do
not fire on. Deliberately medium/low-confidence because benign rule changes are
common; the escalation is an inbox-rule change from a client IP the offline IOC /
reputation sweep has flagged, or from an unfamiliar IP right after a suspicious
sign-in — correlate this alert's `source_ip` against those findings.
match:
provider: Office365
event_action_any:
- New-InboxRule
- Set-InboxRule
- UpdateInboxRules
- Remove-InboxRule
- Enable-InboxRule
- Disable-InboxRule
correlation_keys: [user_name, source_ip]
false_positive_guidance: |
Users and admins create, edit, and delete their own inbox rules constantly, so
this fires broadly and is a LEAD, not a verdict — expect to tune it per tenant.
The high-fidelity signal is an inbox-rule change whose `source_ip` matches the
case's IP/IOC blocklist sweep or a suspicious sign-in for the same mailbox, or a
forwarding/deletion rule (already raised at high severity by the dedicated rules).
Confirm the mailbox owner made the change and read the rule name/actions in the
event evidence before escalating.
Related detections
Run this rule on your own logs — offline
KIROSEC runs this rule and the rest of the library over Windows, Linux, macOS and cloud audit logs on the analyst's own machine. Every alert links to the events that raised it, and nothing leaves the device.
