Cross-Cloud Suspicious Compute resource creation in GCP

Severity: low
Time window: 1d
Group by: AWSNetworkEntity, GCPUserIp
Source: github.com/Azure/Azure-Sentinel

'This detection identifies potential suspicious activity across multi-cloud environments by combining AWS GuardDuty findings with GCP Audit Logs. It focuses on AWS activities related to unauthorized access, credential abuse, and unusual behaviors, as well as GCP instances creation with non-Google service account users. The query aims to provide a comprehensive view of cross-cloud security incidents for proactive threat detection and response.'

MITRE ATT&CK coverage

Tactic	Techniques
Initial Access	`T1078` Valid Accounts, `T1566` Phishing
Execution	`T1059` Command and Scripting Interpreter
Persistence	`T1078` Valid Accounts, `T1547` Boot or Logon Autostart Execution
Privilege Escalation	`T1078` Valid Accounts, `T1547` Boot or Logon Autostart Execution, `T1548` Abuse Elevation Control Mechanism
Credential Access	`T1552` Unsecured Credentials
Discovery	`T1069` Permission Groups Discovery

Rules detecting the same action

Other rules on this platform that filter on the same API call or operation.

Rule body kusto

id: 5c847e47-0a07-4c01-ab99-5817ad6cb11e
name: Cross-Cloud Suspicious Compute resource creation in GCP
description: |
  'This detection  identifies potential suspicious activity across multi-cloud environments by combining AWS GuardDuty findings with GCP Audit Logs. It focuses on AWS activities related to unauthorized access, credential abuse, and unusual behaviors, as well as GCP instances creation with non-Google service account users. The query aims to provide a comprehensive view of cross-cloud security incidents for proactive threat detection and response.'
severity: Low
requiredDataConnectors:
  - connectorId: GCPAuditLogsDefinition
    dataTypes:
      - GCPAuditLogs
  - connectorId: AWSS3
    dataTypes:
      - AWSGuardDuty
queryFrequency: 1d
queryPeriod: 1d
triggerOperator: gt
triggerThreshold: 0
tactics:
  - InitialAccess
  - Execution
  - Persistence
  - PrivilegeEscalation
  - CredentialAccess
  - Discovery
  - LateralMovement
relevantTechniques:
  - T1566
  - T1059
  - T1078
  - T1547
  - T1548
  - T1069
  - T1552
query: |
  // Materialize AWS GuardDuty findings
  let AwsAlert = materialize (
      AWSGuardDuty
      // Filter for specific activity types in AWS GuardDuty
      | where ActivityType has_any (
          "Backdoor:EC2/DenialOfService.UnusualProtocol",
          "CredentialAccess:Kubernetes/MaliciousIPCaller",
          "CredentialAccess:Kubernetes/SuccessfulAnonymousAccess",
          "CredentialAccess:Kubernetes/TorIPCaller",
          "CredentialAccess:RDS/AnomalousBehavior.SuccessfulBruteForce",
          "CredentialAccess:RDS/TorIPCaller.FailedLogin",
          "CredentialAccess:RDS/TorIPCaller.SuccessfulLogin",
          "Discovery:Kubernetes/MaliciousIPCaller",
          "Recon:IAMUser/MaliciousIPCaller.Custom",
          "UnauthorizedAccess:EC2/TorClient",
          "UnauthorizedAccess:IAMUser/TorIPCaller",
          "UnauthorizedAccess:IAMUser/MaliciousIPCaller.Custom",
          "UnauthorizedAccess:IAMUser/InstanceCredentialExfiltration.OutsideAWS",
          "UnauthorizedAccess:IAMUser/InstanceCredentialExfiltration.InsideAWS",
          "UnauthorizedAccess:IAMUser/ConsoleLoginSuccess.B"
          )
      // Extract and transform AWS GuardDuty attributes
      | extend
          AWSAlertId = Id, 
          AWSAlertTitle = Title,
          AWSAlertDescription = Description,
          AWSresourceType = tostring(parse_json(ResourceDetails).resourceType),
          AWSNetworkEntity = tostring(parse_json(ServiceDetails).action.awsApiCallAction.remoteIpDetails.ipAddressV4),
          AWSAlertUserNameEntity = tostring(parse_json(ResourceDetails).accessKeyDetails.userName),
          InstanceType = tostring(parse_json(ResourceDetails).instanceDetails.instanceType),
          AWSTargetingService = parse_json(ServiceDetails).additionalInfo.apiCalls,
          AWSAlertTime = TimeCreated,
          AWSAlertLink= tostring(strcat('https://us-east-1.console.aws.amazon.com/guardduty/home?region=us-east-1#/findings?macros=current&fId=', Id)),
          Severity = 
          case (
      Severity >= 7.0,
      "High",
      Severity between (4.0 .. 6.9),
      "Medium",
      Severity between (1.0 .. 3.9),
      "Low",
      "Unknown"
  )
      // Extract API call details and count
      | mv-apply AIPCall = AWSTargetingService on 
          ( 
          where AIPCall has "name"    
          | extend APICallName = tostring(AIPCall.name), APICallCount = tostring(AIPCall["count"])
          ) 
      // Select distinct attributes for further analysis
      | distinct
          AWSAlertTime,
          ActivityType,
          Severity,
          AWSAlertId,
          AWSAlertTitle,
          AWSAlertDescription,
          AWSAlertLink,
          Arn,
          AWSresourceType,
          AWSNetworkEntity,
          AWSAlertUserNameEntity,
          InstanceType,
          APICallName,
          APICallCount      
      );
  // Materialize GCP Audit Logs related to VM instance creation
  let GCPVMActivity= materialize(
      GCPAuditLogs 
      // Filter for Compute Engine instances insertions
      | where ServiceName == "compute.googleapis.com" and MethodName endswith "instances.insert"
      // Extract and transform relevant GCP Audit Log attributes
      | extend
          GCPUserUPN= tostring(parse_json(AuthenticationInfo).principalEmail),
          GCPUserIp = tostring(parse_json(RequestMetadata).callerIp),
          GCPUserUA= tostring(parse_json(RequestMetadata).callerSuppliedUserAgent),
          VMDetails= parse_json(AuthorizationInfo),
          VMStatus =  tostring(parse_json(Response).status),
          VMOperation=tostring(parse_json(Response).operationType),
          VMName= tostring(parse_json(Request).name),
          VMDescription= tostring(parse_json(Request).description),
          VMType = tostring(split(parse_json(Request).machineType, "/")[-1]),
          Tags= tostring(parse_json(Request).tags),
          RequestJS = parse_json(Request)
      // Filter out service account-related activities and private IP addresses
      | where GCPUserUPN !has "gserviceaccount.com"
      | extend Name = tostring(split(GCPUserUPN, "@")[0]), UPNSuffix = tostring(split(GCPUserUPN, "@")[1])
      | where VMOperation == "insert" and isnotempty(GCPUserIp) and GCPUserIp != "private"
      // Select relevant attributes for further analysis
      | project
          GCPOperationTime=TimeGenerated,
          VMName,
          VMStatus,
          MethodName,
          GCPUserUPN,
          ProjectId,
          GCPUserIp,
          GCPUserUA,
          VMOperation,
          VMType,
          Name,
          UPNSuffix
      );
  // Join AWS and GCP activities based on matching IP addresses
  AwsAlert
  | join kind= inner (GCPVMActivity)
      on
      $left.AWSNetworkEntity == $right.GCPUserIp
entityMappings:
  - entityType: IP
    fieldMappings:
      - identifier: Address
        columnName: GCPUserIp
  - entityType: Account
    fieldMappings:
      - identifier: FullName
        columnName: GCPUserUPN
      - identifier: Name
        columnName: Name
      - identifier: UPNSuffix
        columnName: UPNSuffix
customDetails:
  AWSAlertUserName: AWSAlertUserNameEntity
  AWSArn: Arn
  AWSresourceType: AWSresourceType
  AWSInstanceType: InstanceType
  AWSAPICallName: APICallName
  AWSAPICallCount: APICallCount
  GCPUserAgent: GCPUserUA
  GCPVMName: VMName
  GCPProjectId: ProjectId
  GCPVMType: VMType
  CorrelationWith: "GCPAuditLogs"
alertDetailsOverride:
  alertDisplayNameFormat: "{{AWSNetworkEntity}} from {{AWSAlertTitle}} observed in GCP compute activity with {{GCPUserUPN}}"
  alertDescriptionFormat: " This detection compiles and correlates unauthorized user access alerts originating from AWS GuardDuty With Alert Description '{{AWSAlertDescription}}' assocated with GCP compute activities. It focuses on AWS GuardDuty alerts related to unauthorized user access, specifically targeting network IP associations tied to activities such as logins from malicious IP addresses or instance credential exfiltration attempts. The detection leverages these common network IP advisories to detect and pinpoint unauthorized users attempting to access both AWS and Azure resources.  \n\n AWS ALert Link : '{{AWSAlertLink}}' \n\n Find More Details :https://docs.aws.amazon.com/guardduty/latest/ug/guardduty_finding-types-active.html"
  alertSeverityColumnName: Severity
  alertDynamicProperties:
    - alertProperty: AlertLink
      value: AWSAlertLink
    - alertProperty: ProviderName
      value: "AWS"
    - alertProperty: ProductComponentName
      value: "AWSGuarduty"
kind: Scheduled
version: 1.0.2

Stages and Predicates

Let binding: `GCPVMActivity`

let GCPVMActivity = materialize(
    GCPAuditLogs 
    | where ServiceName == "compute.googleapis.com" and MethodName endswith "instances.insert"
    | extend
        GCPUserUPN= tostring(parse_json(AuthenticationInfo).principalEmail),
        GCPUserIp = tostring(parse_json(RequestMetadata).callerIp),
        GCPUserUA= tostring(parse_json(RequestMetadata).callerSuppliedUserAgent),
        VMDetails= parse_json(AuthorizationInfo),
        VMStatus =  tostring(parse_json(Response).status),
        VMOperation=tostring(parse_json(Response).operationType),
        VMName= tostring(parse_json(Request).name),
        VMDescription= tostring(parse_json(Request).description),
        VMType = tostring(split(parse_json(Request).machineType, "/")[-1]),
        Tags= tostring(parse_json(Request).tags),
        RequestJS = parse_json(Request)
    | where GCPUserUPN !has "gserviceaccount.com"
    | extend Name = tostring(split(GCPUserUPN, "@")[0]), UPNSuffix = tostring(split(GCPUserUPN, "@")[1])
    | where VMOperation == "insert" and isnotempty(GCPUserIp) and GCPUserIp != "private"
    | project
        GCPOperationTime=TimeGenerated,
        VMName,
        VMStatus,
        MethodName,
        GCPUserUPN,
        ProjectId,
        GCPUserIp,
        GCPUserUA,
        VMOperation,
        VMType,
        Name,
        UPNSuffix
    );

The stages below define let AwsAlert (the rule's main pipeline source).

Stage 1: `source`

AWSGuardDuty

Stage 2: `where`

| where ActivityType has_any (
        "Backdoor:EC2/DenialOfService.UnusualProtocol",
        "CredentialAccess:Kubernetes/MaliciousIPCaller",
        "CredentialAccess:Kubernetes/SuccessfulAnonymousAccess",
        "CredentialAccess:Kubernetes/TorIPCaller",
        "CredentialAccess:RDS/AnomalousBehavior.SuccessfulBruteForce",
        "CredentialAccess:RDS/TorIPCaller.FailedLogin",
        "CredentialAccess:RDS/TorIPCaller.SuccessfulLogin",
        "Discovery:Kubernetes/MaliciousIPCaller",
        "Recon:IAMUser/MaliciousIPCaller.Custom",
        "UnauthorizedAccess:EC2/TorClient",
        "UnauthorizedAccess:IAMUser/TorIPCaller",
        "UnauthorizedAccess:IAMUser/MaliciousIPCaller.Custom",
        "UnauthorizedAccess:IAMUser/InstanceCredentialExfiltration.OutsideAWS",
        "UnauthorizedAccess:IAMUser/InstanceCredentialExfiltration.InsideAWS",
        "UnauthorizedAccess:IAMUser/ConsoleLoginSuccess.B"
        )

Stage 3: `extend`

| extend
        AWSAlertId = Id, 
        AWSAlertTitle = Title,
        AWSAlertDescription = Description,
        AWSresourceType = tostring(parse_json(ResourceDetails).resourceType),
        AWSNetworkEntity = tostring(parse_json(ServiceDetails).action.awsApiCallAction.remoteIpDetails.ipAddressV4),
        AWSAlertUserNameEntity = tostring(parse_json(ResourceDetails).accessKeyDetails.userName),
        InstanceType = tostring(parse_json(ResourceDetails).instanceDetails.instanceType),
        AWSTargetingService = parse_json(ServiceDetails).additionalInfo.apiCalls,
        AWSAlertTime = TimeCreated,
        AWSAlertLink= tostring(strcat('https://us-east-1.console.aws.amazon.com/guardduty/home?region=us-east-1#/findings?macros=current&fId=', Id)),
        Severity = 
        case (
    Severity >= 7.0,
    "High",
    Severity between (4.0 .. 6.9),
    "Medium",
    Severity between (1.0 .. 3.9),
    "Low",
    "Unknown"
)

Severity =

ifSeverity >= 7.0"High"

elifSeverity >= 4.0 and Severity <= 6.9"Medium"

elifSeverity >= 1.0 and Severity <= 3.9"Low"

else"Unknown"

Stage 4: `kusto:mv-apply`

| mv-apply AIPCall = AWSTargetingService on 
        ( 
        where AIPCall has "name"    
        | extend APICallName = tostring(AIPCall.name), APICallCount = tostring(AIPCall["count"])
        )

Stage 5: `distinct`

| distinct
        AWSAlertTime,
        ActivityType,
        Severity,
        AWSAlertId,
        AWSAlertTitle,
        AWSAlertDescription,
        AWSAlertLink,
        Arn,
        AWSresourceType,
        AWSNetworkEntity,
        AWSAlertUserNameEntity,
        InstanceType,
        APICallName,
        APICallCount

The stages below run on AwsAlert (the outer pipeline).

Stage 6: `join`

AwsAlert
| join kind= inner (GCPVMActivity)
    on
    $left.AWSNetworkEntity == $right.GCPUserIp

Exclusions

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

Field	Kind	Excluded values
`GCPUserUPN`	match	`gserviceaccount.com`

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
`AIPCall`	match	`name` transforms: `term`
`ActivityType`	match	`Backdoor:EC2/DenialOfService.UnusualProtocol` `CredentialAccess:Kubernetes/MaliciousIPCaller` `CredentialAccess:Kubernetes/SuccessfulAnonymousAccess` `CredentialAccess:Kubernetes/TorIPCaller` `CredentialAccess:RDS/AnomalousBehavior.SuccessfulBruteForce` `CredentialAccess:RDS/TorIPCaller.FailedLogin` `CredentialAccess:RDS/TorIPCaller.SuccessfulLogin` `Discovery:Kubernetes/MaliciousIPCaller` `Recon:IAMUser/MaliciousIPCaller.Custom` `UnauthorizedAccess:EC2/TorClient` `UnauthorizedAccess:IAMUser/ConsoleLoginSuccess.B` `UnauthorizedAccess:IAMUser/InstanceCredentialExfiltration.InsideAWS` `UnauthorizedAccess:IAMUser/InstanceCredentialExfiltration.OutsideAWS` `UnauthorizedAccess:IAMUser/MaliciousIPCaller.Custom` `UnauthorizedAccess:IAMUser/TorIPCaller`
`GCPUserIp`	is_not_null	(no value, null check)
`GCPUserIp`	ne	`private` transforms: `cased`
`MethodName`	ends_with	`instances.insert`
`ServiceName`	eq	`compute.googleapis.com` transforms: `cased`
`VMOperation`	eq	`insert` transforms: `cased`

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
`AWSAlertDescription`	`extend`
`AWSAlertId`	`extend`
`AWSAlertLink`	`extend`
`AWSAlertTime`	`extend`
`AWSAlertTitle`	`extend`
`AWSAlertUserNameEntity`	`extend`
`AWSNetworkEntity`	`extend`
`AWSTargetingService`	`extend`
`AWSresourceType`	`extend`
`InstanceType`	`extend`
`Severity`	`extend`

`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)

Cross-Cloud Suspicious Compute resource creation in GCP

MITRE ATT&CK coverage

Rules detecting the same action

Rule body kusto

Stages and Predicates

Let binding: `GCPVMActivity`

Stage 1: `source`

Stage 2: `where`

Stage 3: `extend`

Stage 4: `kusto:mv-apply`

Stage 5: `distinct`

Stage 6: `join`

Exclusions

Indicators

Output fields

Keyboard shortcuts

Search operators

Cross-Cloud Suspicious Compute resource creation in GCP

MITRE ATT&CK coverage

Rules detecting the same action

Rule body kusto

Stages and Predicates

Let binding: GCPVMActivity

Stage 1: source

Stage 2: where

Stage 3: extend

Stage 4: kusto:mv-apply

Stage 5: distinct

Stage 6: join

Exclusions

Indicators

Output fields

Let binding: `GCPVMActivity`

Stage 1: `source`

Stage 2: `where`

Stage 3: `extend`

Stage 4: `kusto:mv-apply`

Stage 5: `distinct`

Stage 6: `join`