Work with STIX objects and indicators to enhance threat intelligence and threat hunting in Microsoft Sentinel (Preview)

On April 3, 2025, we publicly previewed two new tables to support STIX (Structured Threat Information eXpression) indicator and object schemas: ThreatIntelIndicators and ThreatIntelObjects. This article provides examples of how to incorporate STIX objects into queries to enhance threat hunting, and how to migrate to the new threat indicator schema.

For more information about threat intelligence in Microsoft Sentinel, see Threat intelligence in Microsoft Sentinel.

Identify threat actors associated with specific threat indicators

This query is an example of how to correlate threat indicators, such as IP addresses, with threat actors:

 let IndicatorsWithThatIP = (ThreatIntelIndicators
| extend tlId = tostring(Data.id)
| summarize arg_max(TimeGenerated,*) by Id
|  where IsDeleted == false);
let ThreatActors = (ThreatIntelObjects
| where StixType == 'threat-actor'
| extend tlId = tostring(Data.id)
| extend ThreatActorName = Data.name
| extend ThreatActorSource = base64_decode_tostring(tostring(split(Id, '---')[0]))
| summarize arg_max(TimeGenerated,*) by Id
|  where IsDeleted == false);
let AllRelationships = (ThreatIntelObjects
| where StixType == 'relationship'
| extend tlSourceRef = tostring(Data.source_ref)
| extend tlTargetRef = tostring(Data.target_ref)
| extend tlId = tostring(Data.id)
| summarize arg_max(TimeGenerated,*) by Id
|  where IsDeleted == false);
let IndicatorAsSource = (IndicatorsWithThatIP
| join AllRelationships on $left.tlId == $right.tlSourceRef
| join ThreatActors on $left.tlTargetRef == $right.tlId);
let IndicatorAsTarget = (IndicatorsWithThatIP
| join AllRelationships on $left.tlId == $right.tlTargetRef
| join ThreatActors on $left.tlSourceRef == $right.tlId);
IndicatorAsSource
| union IndicatorAsTarget
| project ObservableValue, ThreatActorName

List threat intelligence data related to a specific threat actor

This query provides insights into the tactics, techniques, and procedures (TTPs) of the threat actor (replace Sangria Tempest with the name of the threat actor you want to investigate):

let THREAT_ACTOR_NAME = 'Sangria Tempest';
let ThreatIntelObjectsPlus = (ThreatIntelObjects
| union (ThreatIntelIndicators
| extend StixType = 'indicator')
| extend tlId = tostring(Data.id)
| extend PlusStixTypes = StixType
| extend importantfield = case(StixType == "indicator", Data.pattern,
                            StixType == "attack-pattern", Data.name,
                            "Unkown")
| extend feedSource = base64_decode_tostring(tostring(split(Id, '---')[0]))
| summarize arg_max(TimeGenerated,*) by Id
|  where IsDeleted == false);
let ThreatActorsWithThatName = (ThreatIntelObjects
| where StixType == 'threat-actor'
| where Data.name == THREAT_ACTOR_NAME
| extend tlId = tostring(Data.id)
| extend ActorName = tostring(Data.name)
| summarize arg_max(TimeGenerated,*) by Id
|  where IsDeleted == false);
let AllRelationships = (ThreatIntelObjects
| where StixType == 'relationship'
| extend tlSourceRef = tostring(Data.source_ref)
| extend tlTargetRef = tostring(Data.target_ref)
| extend tlId = tostring(Data.id)
| summarize arg_max(TimeGenerated,*) by Id
|  where IsDeleted == false);
let SourceRelationships = (ThreatActorsWithThatName
| join AllRelationships on $left.tlId == $right.tlSourceRef
| join ThreatIntelObjectsPlus on $left.tlTargetRef == $right.tlId);
let TargetRelationships = (ThreatActorsWithThatName
| join AllRelationships on $left.tlId == $right.tlTargetRef
| join ThreatIntelObjectsPlus on $left.tlSourceRef == $right.tlId);
SourceRelationships
| union TargetRelationships
| project ActorName, PlusStixTypes, ObservableValue, importantfield, Tags, feedSource

Migrate existing queries to the new ThreatIntelObjects schema

This example shows how to migrate existing queries from the legacy ThreatIntelligenceIndicator table to the new ThreatIntelObjects schema. The query uses the extend operator to recreate legacy columns based on the ObservableKey and ObservableValue columns in the new table.

ThreatIntelIndicators
| extend NetworkIP = iff(ObservableKey == 'ipv4-addr:value', ObservableValue, ''),
        NetworkSourceIP = iff(ObservableKey == 'network-traffic:src_ref.value', ObservableValue, ''),
        NetworkDestinationIP = iff(ObservableKey == 'network-traffic:dst_ref.value', ObservableValue, ''),
        DomainName = iff(ObservableKey == 'domain-name:value', ObservableValue, ''),
        EmailAddress = iff(ObservableKey == 'email-addr:value', ObservableValue, ''),
        FileHashType = case(ObservableKey has 'MD5', 'MD5',
                                ObservableKey has 'SHA-1', 'SHA-1',
                                ObservableKey has 'SHA-256', 'SHA-256',
                                ''),
        FileHashValue = iff(ObservableKey has 'file:hashes', ObservableValue, ''),
        Url = iff(ObservableKey == 'url:value', ObservableValue, ''),
        x509Certificate = iff(ObservableKey has 'x509-certificate:hashes.', ObservableValue, ''),
        x509Issuer = iff(ObservableKey has 'x509-certificate:issuer', ObservableValue, ''),
        x509CertificateNumber = iff(ObservableKey == 'x509-certificate:serial_number', ObservableValue, ''),        
        Description = tostring(Data.description),
        CreatedByRef = Data.created_by_ref,
        Extensions = Data.extensions,
        ExternalReferences = Data.references,
        GranularMarkings = Data.granular_markings,
        IndicatorId = tostring(Data.id),
        ThreatType = tostring(Data.indicator_types[0]),
        KillChainPhases = Data.kill_chain_phases,
        Labels = Data.labels,
        Lang = Data.lang,
        Name = Data.name,
        ObjectMarkingRefs = Data.object_marking_refs,
        PatternType = Data.pattern_type,
        PatternVersion = Data.pattern_version,
        Revoked = Data.revoked,
        SpecVersion = Data.spec_version
| project-reorder TimeGenerated, WorkspaceId, AzureTenantId, ThreatType, ObservableKey, ObservableValue, Confidence, Name, Description, LastUpdateMethod, SourceSystem, Created, Modified, ValidFrom, ValidUntil, IsDeleted, Tags, AdditionalFields, CreatedByRef, Extensions, ExternalReferences, GranularMarkings, IndicatorId, KillChainPhases, Labels, Lang, ObjectMarkingRefs, Pattern, PatternType, PatternVersion, Revoked, SpecVersion, NetworkIP, NetworkDestinationIP, NetworkSourceIP, DomainName, EmailAddress, FileHashType, FileHashValue, Url, x509Certificate, x509Issuer, x509CertificateNumber, Data

Related content

For more information, see the following articles:

• Threat intelligence in Microsoft Sentinel.
• Connect Microsoft Sentinel to STIX/TAXII threat intelligence feeds.
• See which TIPs, TAXII feeds, and enrichments can be readily integrated with Microsoft Sentinel.

For more information on KQL, see Kusto Query Language (KQL) overview.

Other resources:

• KQL quick reference
• Kusto Query Language learning resources