Audit Adminactionsfromriskyusers

KQL detection rule by reprise99 · MIT

Finds Azure AD audit events from users who hold admin privileges (based on PIM activations in the last 60 days) and then finds any audit events events from those users in the last 7 days that have elevated risk associated to them This query is part of The Definitive Guide to KQL: Using Kusto Query Language for Operations, Defending, and Threat Hunting - https://aka.ms/KQLMSPress and was contributed by Corissa K - https://www.linkedin.com/in/corissakoopmans/

//Finds Azure AD audit events from users who hold admin privileges (based on PIM activations in the last 60 days) and then finds any audit events events from those users in the last 7 days that have elevated risk associated to them
//This query is part of The Definitive Guide to KQL: Using Kusto Query Language for Operations, Defending, and Threat Hunting - https://aka.ms/KQLMSPress and was contributed by Corissa K - https://www.linkedin.com/in/corissakoopmans/

//Data connector required for this query - Azure Active Directory - Audit Logs

let privroles = pack_array("Application Administrator","Authentication Administrator","Cloud Application Administrator","Conditional Access Administrator","Exchange Administrator","Global Administrator","Helpdesk Administrator","Hybrid Identity Administrator","Password Administrator","Privileged Authentication Administrator","Privileged Role Administrator","Security Administrator","SharePoint Administrator","User Administrator");
let privusers = AuditLogs 
| where TimeGenerated > ago(60d) and ActivityDisplayName == 'Add member to role completed (PIM activation)' and Category == "RoleManagement" 
| extend Caller = tostring(InitiatedBy.user.userPrincipalName) 
| extend Role = tostring(TargetResources[0].displayName) 
| where Role in (privroles) 
| distinct Caller;
let Activity =  AuditLogs
    | mv-expand ParsedFields = parse_json(TargetResources)
    | extend Target = tostring(ParsedFields.userPrincipalName), DisplayName = tostring(ParsedFields.displayName)
    | project TimeGenerated, Target, DisplayName, ParsedFields, OperationName;
    let RiskyUsers = SigninLogs
    | where RiskLevelDuringSignIn == "high"
    | where RiskState == "atRisk"
    | project TimeGenerated,UserPrincipalName, UserDisplayName, RiskDetail, RiskLevelDuringSignIn, RiskState;
    Activity
    | join kind=inner(RiskyUsers) on $left.DisplayName==$right.UserDisplayName
    | where TimeGenerated >= ago(7d) and UserPrincipalName in~ (privusers)
    | distinct UserDisplayName, RiskDetail, RiskLevelDuringSignIn, OperationName

[[ currentRule.title && currentRule.title.length > 60 ? currentRule.title.slice(0, 60) + '…' : currentRule.title ]]

dfc61633-2f5c-472f-b770-ec595c345041

Propose edits, view related rules, and engage with the community through comments.

Executable code, not a declarative rule

This submission is arbitrary [[ currentRule.format ? currentRule.format.toUpperCase() : '' ]] code, not a declarative detection rule — review it as you would any script before running it, and note that it cannot be safely aggregated or auto-executed alongside YARA/Sigma/Suricata/CRS rules.

Rejected pattern present

[[ reason ]]

What does this mean?

Format & documentation suggestions

Not a security issue — this rule is valid and was accepted as-is. These are optional fields/conventions its own format recommends, mainly affecting its quality score and how well it documents itself for other reviewers.

[[ reason ]]

What do these messages mean?

[[ currentRule.title && currentRule.title.length > 60 ? currentRule.title.slice(0, 60) + '…' : currentRule.title ]]

[[ currentRule.title ]]

Community Discussion