Azure Protection Multiple Alerts for User

Severity: high
Group by: properties.userPrincipalName
Log types: Azure.Audit
Reference: https://github.com/elastic/detection-rules/blob/main/rules/integrations/azure/initial_access_entra_id_protection_alerts_for_user.toml
Source: github.com/panther-labs/panther-analysis

Detects when a user account triggers multiple Microsoft Entra ID Protection risk alerts within a short time window, indicating a potentially ongoing attack or compromised account. Entra ID Protection uses machine learning and heuristics to detect risky sign-in activity including anonymous IP usage, atypical travel patterns, malware-linked IP addresses, unfamiliar sign-in properties, password spray attempts, leaked credentials, and other suspicious behaviors.

MITRE ATT&CK coverage

Tactic	Techniques
Initial Access	`T1078.004` Valid Accounts: Cloud Accounts

Rule body yaml

AnalysisType: rule
Filename: azure_multiple_protection_alerts.py
RuleID: "Azure.Audit.MultipleProtectionAlerts"
DisplayName: "Azure Protection Multiple Alerts for User"
Enabled: true
LogTypes:
  - Azure.Audit
Severity: High
Threshold: 3
DedupPeriodMinutes: 15
Description: >
  Detects when a user account triggers multiple Microsoft Entra ID Protection risk alerts within
  a short time window, indicating a potentially ongoing attack or compromised account. Entra ID
  Protection uses machine learning and heuristics to detect risky sign-in activity including
  anonymous IP usage, atypical travel patterns, malware-linked IP addresses, unfamiliar sign-in
  properties, password spray attempts, leaked credentials, and other suspicious behaviors.
Reports:
  MITRE ATT&CK:
    - TA0001:T1078
    - TA0001:T1078.004
Runbook: |
  1. Query Azure.Audit logs for all sign-in events by properties:userPrincipalName in the 1 hour surrounding this alert to identify the full sequence of risk events, noting properties:riskEventTypes, callerIpAddress, properties:location, and properties:userAgent for each authentication attempt
  2. Review properties:riskDetail and properties:riskState for each event to understand the specific risk detections that triggered the alerts, and check if multiple different risk types occurred (e.g., anonymous IP + atypical travel + unfamiliar properties) which strongly indicates account compromise rather than false positives
  3. Immediately disable the user account or revoke all active sessions if suspicious activity is confirmed, force password reset with MFA re-enrollment, review Azure.Audit logs for any API calls or resource access by the user in the 24 hours after the first risk event to identify data exfiltration or privilege escalation, and contact the user through out-of-band communication to verify if they are aware of the authentication attempts
Reference: https://github.com/elastic/detection-rules/blob/main/rules/integrations/azure/initial_access_entra_id_protection_alerts_for_user.toml
SummaryAttributes:
  - properties:userPrincipalName
  - callerIpAddress
  - properties:riskState
  - properties:riskLevelDuringSignIn
Tests:
  - Name: Risk Event Counts Toward Threshold
    ExpectedResult: true
    Log:
      {
        "time": "2025-01-15 09:35:45.789",
        "resourceId": "/tenants/tenant-123/providers/Microsoft.aadiam",
        "operationName": "Sign-in activity",
        "operationVersion": "1.0",
        "category": "SignInLogs",
        "tenantId": "tenant-123",
        "resultType": "0",
        "resultSignature": "SUCCESS",
        "durationMs": 0,
        "callerIpAddress": "3.3.3.3",
        "correlationId": "risk-event-003",
        "Level": "4",
        "properties":
          {
            "createdDateTime": "2025-01-15T09:35:45.7890123Z",
            "userPrincipalName": "aragorn@lotr.com",
            "userId": "user-victim-123",
            "ipAddress": "3.3.3.3",
            "riskState": "confirmedCompromised",
            "riskLevelDuringSignIn": "high",
            "riskLevelAggregated": "high",
            "riskDetail": "aiConfirmedSigninCompromised",
            "riskEventTypes": ["leakedCredentials", "anomalousToken"],
            "authenticationProtocol": "oAuth2",
            "clientAppUsed": "Browser",
            "isInteractive": true,
            "location": { "city": "Beijing", "countryOrRegion": "China" },
            "deviceDetail": { "browser": "Firefox", "operatingSystem": "Linux" },
            "userAgent": "Mozilla/5.0 (X11; Linux x86_64)",
          },
        "p_event_time": "2025-01-15 09:35:45.789",
        "p_log_type": "Azure.Audit",
        "p_parse_time": "2025-01-15 09:35:46.000",
        "p_row_id": "row-003",
      }
  - Name: No Risk Sign-In
    ExpectedResult: false
    Log:
      {
        "time": "2025-01-15 10:00:30.012",
        "resourceId": "/tenants/tenant-456/providers/Microsoft.aadiam",
        "operationName": "Sign-in activity",
        "operationVersion": "1.0",
        "category": "SignInLogs",
        "tenantId": "tenant-456",
        "resultType": "0",
        "resultSignature": "SUCCESS",
        "durationMs": 0,
        "callerIpAddress": "203.0.113.200",
        "correlationId": "normal-signin-001",
        "Level": "4",
        "properties":
          {
            "createdDateTime": "2025-01-15T10:00:30.0123456Z",
            "userPrincipalName": "normaluser@company.com",
            "userId": "user-normal-456",
            "ipAddress": "203.0.113.200",
            "riskState": "none",
            "riskLevelDuringSignIn": "none",
            "riskLevelAggregated": "none",
            "riskDetail": "none",
            "riskEventTypes": [],
            "authenticationProtocol": "oAuth2",
            "clientAppUsed": "Browser",
            "isInteractive": true,
            "location": { "city": "Seattle", "countryOrRegion": "United States" },
            "deviceDetail": { "browser": "Chrome", "operatingSystem": "Windows 11" },
            "userAgent": "Mozilla/5.0 (Windows NT 10.0)",
          },
        "p_event_time": "2025-01-15 10:00:30.012",
        "p_log_type": "Azure.Audit",
      }

Detection logic

Condition

not (operationName ne "Sign-in activity" or resultSignature ne "SUCCESS")
not (properties.riskState is_null or properties.riskState not in ["atrisk", "confirmedcompromised"])
properties.riskLevelDuringSignIn in ["high", "medium"] or properties.riskLevelAggregated in ["high", "medium"]
properties.userPrincipalName is_not_null

Exclusions

Top-level NOT(...) conjuncts: predicates this rule actively suppresses.

Field	Kind	Excluded values
`properties.riskState`	in	`atrisk`, `confirmedcompromised`
`properties.riskState`	is_null	(no value, null check)
`operationName`	ne	`Sign-in activity`
`resultSignature`	ne	`SUCCESS`

Indicators

Each row is a field, operator, and value that the rule matches. The corpus column counts how many other rules in the catalog look for the same combination: high numbers point to widely-used, community-vetted indicators. Blank or 1 shows that the indicator is specific to this rule.

Field	Kind	Values
`properties.riskLevelAggregated`	in	`high` `medium`
`properties.riskLevelDuringSignIn`	in	`high` `medium`
`properties.riskState`	in	`atrisk` `confirmedcompromised`
`properties.userPrincipalName`	is_not_null	(no value, null check)

Output fields

Fields the rule emits when it matches. Chronicle authors list these in the outcome block; they appear on the detection and $risk_score drives alerting. Sentinel / Defender XDR rules build them up through project / summarize / extend stages. Sentinel maps these into alert fields via entityMappings and customDetails; Defender XDR custom detections surface them as alert fields directly.

Field	Source
`userPrincipalName`	`properties.userPrincipalName`

`j` / `k`	Scroll down / up
`d` / `u`	Half-page down / up
`gg` / `G`	Top / bottom
`h` / `l`	History back / forward
`f`	Follow link (`Shift` = new tab)
`/`	Focus search
`?`	Toggle this help
`↑` / `↓`	Navigate search results
`Enter`	Open highlighted result
`Esc`	Close results / dialog

`type:`	`events` / `rules` / `providers`
`vendor:`	`sigma` / `elastic` / `splunk` / `kusto` / `chronicle` (vendor name alone also works: `sigma:`, `kql:`, `secops:`…)
`tactic:`	TA-id, slug, or name: `credential_access`, `TA0006`
`technique:`	technique or sub-technique ID: `T1003`, `T1003.001` (alias `tech:`)
`severity:`	`critical` / `high` / `medium` / `low` / `informational` (alias `sev:`)
`risk_score`	Numeric comparison on the Elastic risk score (0 to 100): `risk_score>50`, `risk_score<=20`, `risk_score=99` (alias `risk`; Elastic rules only)
`stages:`	Rules with exactly N pipeline stages
`correlation:`	`single_event` / `sequence` / `alternatives` / `alternatives_cross_log` / `all_required` / `correlated`
`with:`	Co-occurrence event-id; stacks (`with:4624 with:4769`) to require all, while a comma list in one occurrence (`with:4624,4769`) is an either-or group. Implies multi-event
`like:`	Structural neighbors of a rule slug (equivalents + subsumption stricter / broader): `like:comsvcs_lsass_memory_dump-splunk-sysmon`
`groupby:`	Entity-grouping substring match against `group_by_keys`: `groupby:user`, `groupby:host`
`uses:`	Rules whose predicate tree touches the field (any kind, any value): `uses:CommandLine`
`excludes:`	Rules with top-level `not()` clauses on the field (FP whitelists): `excludes:ParentImage`
`field:` / `value:`	Predicate search; narrows rule cards to those with a matching leaf and drives the indicator tier. Unquoted = substring, wildcards allowed (`value:mimikatz`)
`indicator:`	Shorthand for `field:F value:V`: `indicator:Image=*\powershell.exe`
`kind:`	Filter by predicate kind. Narrows rule cards to those carrying a matching predicate leaf (`vendor:elastic kind:cidr_match`) and drives the indicator tier: `contains` / `starts_with` / `ends_with` / `regex` / `cidr` / `eq` / `in` … (operator aliases `op:`/`match:`)
`has:` / `no:`	`sample`, `field`, `notes`, `refs`, `trace`, `thirdparty`, `rule`, `pattern`, `timewindow`, `threshold`, `newterms`, `sigma`/`elastic`/`splunk`/`kusto`/`chronicle`
`-op:val`	Exclude matches; works on most operators but not `type:`/`like:`/`has:`/`no:` (use `no:<flag>` to exclude a rule flag): `tactic:execution -vendor:splunk`. Standalone `-kind:`/`-field:`/`-value:` drop every rule carrying a matching predicate leaf (`type:rules -kind:is_null`)
`field:"…"` / `value:"…"`	Quoted value = anchored exact match (also allows spaces): `value:"net user"`
`a,b`	Comma = OR inside one operator (`vendor:sigma,elastic`, `severity:high,critical`); repeating a facet merges the same way. `field:`/`value:` never split (literal commas)
`vendors:` / `stage:`	Singular and plural spellings fold to the canonical operator and value: `tactics:` = `tactic:`, `type:event` = `type:events`, `correlation:sequences` = `correlation:sequence`, `has:thresholds` = `has:threshold`
`"quoted phrase"`	Exact-match a multi-word phrase (free text)