Detection rules › YARA-L

YARA-L rules: 1s

Entra ID Add User Outside PIM

#
Severity
medium
Type
alert
Time window
5m
Match by
user
Author
Google Cloud Security
Source
github.com/chronicle/detection-rules

Adding users to roles is expected but if it occurs outside of PIM, validation should occur

MITRE ATT&CK Coverage

Telemetry Coverage

Rules Detecting the Same Action

These rules filter on the same operation.

Rule Body

// Copyright 2025 Google LLC. Licensed under Apache-2.0.

 rule entra_id_add_user_outside_pim {

  meta:
    author = "Google Cloud Security"
    description = "Adding users to roles is expected but if it occurs outside of PIM, validation should occur"
    rule_id = "mr_bc0a22e2-7ab1-4368-9408-3e6a5b666bec"
    rule_name = "Entra ID Add User Outside PIM"
    tactic = "TA0003"
    technique = "T1098.003"
    assumption = "The list entra_id_watchlist_roles or entra_id_watchlist_roles_guid is populated with role names, ie Exchange Administrator or 29232cdf-9323-42fd-ade2-1d097af3e4de, respectively"
    type = "alert"
    platform = "azure"
    data_source = "azure ad"
    severity = "Medium"
    priority = "Medium"

  events:
    $app.metadata.event_type = "USER_UNCATEGORIZED"
    $app.metadata.product_name = "Azure AD Directory Audit"
    $app.metadata.vendor_name = "Microsoft"
    $app.security_result.action = "ALLOW"
    $app.metadata.product_event_type = "Add member to role outside of PIM (permanent)" //triggered via PS API call
    //Can add watchlist of specific roles by name using the following commented out line
    //$app.target.resource.name in %entra_id_watchlist_roles
    //Can add watchlist of specific roles by guid using the following commented out line
    //$app.target.resource.attribute.labels["TemplateId"] in %entra_id_watchlist_roles_guid
    $app.principal.user.userid = $user

  match:
    $user over 5m

  outcome:
    $risk_score = 65
    $event_count = count_distinct($app.metadata.id)
    $assigned_role = array_distinct($app.target.resource.name)
    $assigned_role_guid = array_distinct($app.target.resource.attribute.labels["TemplateId"])
    $network_session = array_distinct($app.network.session_id)
    $userid_making_modification = array_distinct($app.principal.user.userid)
    $principal_user_email_addresses = array_distinct($app.principal.user.email_addresses)
    $modified_userid = array_distinct($app.target.user.email_addresses)

  condition:
    $app
}

YARA-L Rule Structure

Condition

Fires when at least one $app event in the 5m window.

Events

$app USER_UNCATEGORIZED

  • security_result.action = "ALLOW"
  • metadata.product_event_type = "Add member to role outside of PIM (permanent)"

Match and Correlation

Match key
$user (principal.user.userid)
Within
5m

Outcome

Outcome variables add context to a detection. When the rule generates an alert, $risk_score is displayed with it.

FieldExpression
risk_score65
event_countcount_distinct($app.metadata.id)
assigned_rolearray_distinct($app.target.resource.name)
assigned_role_guidarray_distinct($app.target.resource.attribute.labels["TemplateId"])
network_sessionarray_distinct($app.network.session_id)
userid_making_modificationarray_distinct($app.principal.user.userid)
principal_user_email_addressesarray_distinct($app.principal.user.email_addresses)
modified_useridarray_distinct($app.target.user.email_addresses)

MITRE ATT&CK T1110.001 Windows Repeated Authentication Failures Before Successful One With User Entity

#
Severity
medium
Type
alert
Time window
15m
Match by
hostname, target_user
Author
Google Cloud Security
Source
github.com/chronicle/detection-rules

Detect Multiple Failed Login Attempts followed by Successful Login

MITRE ATT&CK Coverage

TacticTechniques
Credential Access

Telemetry Coverage

Rule Body

// Copyright 2023 Google LLC. Licensed under Apache-2.0.

rule win_repeatedAuthFailure_thenSuccess_T1110_001_user_asset_entity {

  meta:
    author = "Google Cloud Security"
    description = "Detect Multiple Failed Login Attempts followed by Successful Login"
    rule_id = "mr_2a9dc1d2-d4b8-4b80-a494-3369c681da60"
    rule_name = "MITRE ATT&CK T1110.001 Windows Repeated Authentication Failures Before Successful One With User Entity"
    type = "alert"
    tags = "user entity, asset entity"
    assumption = "Assumes ingestion of Windows users and assets to entity graph to entity graph"
    data_source = "microsoft windows events"
    platform = "Windows"
    severity = "Medium"
    priority = "Medium"
    tactic = "TA0006"
    technique = "T1110.001"

  events:
    $fail.metadata.event_type = "USER_LOGIN"
    $fail.metadata.vendor_name = "Microsoft"
    $fail.principal.hostname = $hostname
    $fail.target.user.userid = $target_user
    $fail.security_result.action = "BLOCK"
    $fail.metadata.product_event_type = "4625"

    $fail.metadata.event_timestamp.seconds < $success.metadata.event_timestamp.seconds

    $success.metadata.event_type = "USER_LOGIN"
    $success.metadata.vendor_name = "Microsoft"
    $success.target.user.userid = $target_user
    $success.principal.hostname = $hostname
    $success.security_result.action = "ALLOW"
    $success.metadata.product_event_type = "4624"

    // Correlate with user entity data
    $user.graph.entity.user.userid = $target_user
    $user.graph.metadata.entity_type = "USER"
    $user.graph.metadata.source_type = "ENTITY_CONTEXT"
    // Can modify the following line to focus the rule on different user groups
    any $user.graph.relations.entity.group.group_display_name = "Domain Admins"

    // Correlate with asset entity data
    $asset.graph.entity.asset.hostname = $hostname
    $asset.graph.metadata.entity_type = "ASSET"
    $asset.graph.metadata.source_type = "ENTITY_CONTEXT"
    // Can modify the following line to focus the rule on different computer groups
    any $asset.graph.relations.entity.group.group_display_name = "Domain Controllers"

  match:
    $target_user, $hostname over 15m

    outcome:
    //increment risk score if the system targeted is active directory server, can modify as you see fit, using this as an example
    $risk_score = max(if($hostname = /activedir/, 75, 50))
    $failed_login_threshold = 5
    $impacted_systems = array_distinct($hostname)
    $impacted_users = array_distinct($target_user)
    $alert_type = "Successful Brute Force Attack"
    $tlp = "red"
    // added to populate alert graph with additional context
    // Commented out principal.hostname because it is already represented in graph as match variable. If match changes, can uncomment to add to results
    //$principal_hostname = array_distinct($success.principal.hostname)
    $principal_ip = array_distinct($success.principal.ip)
    $target_hostname = array_distinct($success.target.hostname)
    $src_hostname = array_distinct($success.src.hostname)
    $target_ip = array_distinct($success.target.ip)
    $principal_user_userid = array_distinct($success.principal.user.userid)
    // Commented out target.user.userid because it is already represented in graph as match variable. If match changes, can uncomment to add to results
    //$target_user_userid = array_distinct($success.target.user.userid)
    $principal_resource_name = array_distinct($success.principal.resource.name)
    $target_resource_name = array_distinct($success.target.resource.name)
    $target_url = array_distinct($success.target.url)

  condition:
    #fail > 4 and $success and $user and $asset
}

YARA-L Rule Structure

Condition

Fires when $fail occurs more than 4 times in the 15m window and at least one $success event in the 15m window and $user matches entity context and $asset matches entity context.

Events

$asset entity context (ASSET)

  • graph.metadata.entity_type = "ASSET"
  • graph.metadata.source_type = "ENTITY_CONTEXT"
  • graph.relations.entity.group.group_display_name = "Domain Controllers"

$fail USER_LOGIN (4625)

  • security_result.action = "BLOCK"
  • metadata.product_event_type = "4625"

$success USER_LOGIN (4624)

  • security_result.action = "ALLOW"
  • metadata.product_event_type = "4624"

$user entity context (USER)

  • graph.metadata.entity_type = "USER"
  • graph.metadata.source_type = "ENTITY_CONTEXT"
  • graph.relations.entity.group.group_display_name = "Domain Admins"

Match and Correlation

Order
$fail before $success
Match key
$hostname (principal.hostname), $target_user (target.user.userid)
Within
15m

Outcome

Outcome variables add context to a detection. When the rule generates an alert, $risk_score is displayed with it.

FieldExpression
risk_scoremax(if($hostname = /activedir/, 75, 50))
failed_login_threshold5
impacted_systemsarray_distinct($hostname)
impacted_usersarray_distinct($target_user)
principal_iparray_distinct($success.principal.ip)
target_hostnamearray_distinct($success.target.hostname)
src_hostnamearray_distinct($success.src.hostname)
target_iparray_distinct($success.target.ip)
principal_user_useridarray_distinct($success.principal.user.userid)
principal_resource_namearray_distinct($success.principal.resource.name)
target_resource_namearray_distinct($success.target.resource.name)
target_urlarray_distinct($success.target.url)

sap hanadb audit trail policy changes

#
Severity
high
Time window
30m
Match by
user
Author
Google Cloud Security
Source
github.com/chronicle/detection-rules

Detects changes to SAP HANA audit policies (Create, Alter, Drop) which could indicate an attempt to evade logging.

MITRE ATT&CK Coverage

TacticTechniques
Defense Impairment

Rule Body

// Copyright 2026 Google LLC. Licensed under Apache-2.0.

rule sap_hanadb_audit_trail_policy_changes {

  meta:
    author = "Google Cloud Security"
    description = "Detects changes to SAP HANA audit policies (Create, Alter, Drop) which could indicate an attempt to evade logging."
    severity = "High"
    tactic = "TA0005"
    technique = "T1562"

  events:
    $e.metadata.product_name = "SAP_HANA_AUDIT"

    $statement = $e.additional.fields["executed_statement"]
    $statement = /AUDIT\s+POLICY/ nocase

    $user = $e.principal.user.user_display_name

  match:
    $user over 30m

  outcome:
    $risk_score = max(85)
    $principal_user = array_distinct($user)
    $statements = array_distinct($statement)
    $system_id = array_distinct($e.target.resource.attribute.labels["system_id"])
    $policy_name = array_distinct($e.security_result.rule_name)

  condition:
    $e
}

YARA-L Rule Structure

Condition

Fires when at least one $e event in the 30m window.

Events

$e

  • additional.fields["executed_statement"] matches "AUDIT\s+POLICY"

Match and Correlation

Match key
$user
Within
30m

Outcome

Outcome variables add context to a detection. When the rule generates an alert, $risk_score is displayed with it.

FieldExpression
risk_scoremax(85)
principal_userarray_distinct($user)
statementsarray_distinct($statement)
system_idarray_distinct($e.target.resource.attribute.labels["system_id"])
policy_namearray_distinct($e.security_result.rule_name)

Back to all YARA-L rules