Condividi tramite


Usare gli oggetti e gli indicatori STIX per migliorare l'intelligence sulle minacce e la ricerca delle minacce in Microsoft Sentinel (anteprima)

Il 3 aprile 2025 sono stati visualizzati in anteprima pubblica due nuove tabelle per supportare l'indicatore STIX (Structured Threat Information eXpression) e gli schemi degli oggetti: ThreatIntelIndicators e ThreatIntelObjects. Questo articolo fornisce esempi di come incorporare oggetti STIX in query per migliorare la ricerca delle minacce e come eseguire la migrazione al nuovo schema dell'indicatore di minaccia.

Per altre informazioni sull'intelligence sulle minacce in Microsoft Sentinel, vedere Intelligence sulle minacce in Microsoft Sentinel.

Importante

Microsoft Sentinel inserisce tutte le informazioni sulle minacce nelle nuove ThreatIntelIndicators tabelle e ThreatIntelObjects , continuando a inserire gli stessi dati nella tabella legacy ThreatIntelligenceIndicator fino al 31 luglio 2025.

Assicurarsi di aggiornare le query personalizzate, le regole di analisi e rilevamento, le cartelle di lavoro e l'automazione per usare le nuove tabelle entro il 31 luglio 2025. Dopo questa data, Microsoft Sentinel smetterà di inserire i dati nella tabella legacy ThreatIntelligenceIndicator . Stiamo aggiornando tutte le soluzioni di intelligence per le minacce predefinite nell'hub del contenuto per sfruttare le nuove tabelle. Per altre informazioni sui nuovi schemi di tabella, vedere ThreatIntelIndicators e ThreatIntelObjects.

Identificare gli attori delle minacce associati a indicatori di minaccia specifici

Questa query è un esempio di come correlare gli indicatori di minaccia, ad esempio gli indirizzi IP, con gli attori delle minacce:

 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

Questa query fornisce informazioni dettagliate sulle tattiche, le tecniche e le procedure (TTP) dell'attore di minaccia (sostituire Sangria Tempest con il nome dell'attore di minaccia da analizzare):

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

Eseguire la migrazione di query esistenti al nuovo schema ThreatIntelObjects

In questo esempio viene illustrato come eseguire la migrazione di query esistenti dalla tabella legacy ThreatIntelligenceIndicator al nuovo ThreatIntelObjects schema. La query usa l'operatore extend per ricreare le colonne legacy in base alle colonne ObservableKey e ObservableValue nella nuova tabella.

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

Per altre informazioni, vedere gli articoli seguenti:

Per altre informazioni su KQL, vedere Panoramica del linguaggio di query Kusto (KQL).

Altre risorse: