Detection rules › Kusto

AWSCloudTrail - Privilege escalation via SSM policy

Status
available
Severity
medium
Time window
1d
Source
github.com/Azure/Azure-Sentinel

Detects inline IAM policy updates that grant broad AWS Systems Manager permissions. This can enable command execution and control actions across managed instances, representing a potential privilege escalation route.

MITRE ATT&CK coverage

TacticTechniques
Privilege EscalationT1098.003 Account Manipulation: Additional Cloud Roles

Rules detecting the same action

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

Rule body kusto

id: c668c09f-5a49-43f9-b249-6b89a31ec8fb
name: AWSCloudTrail - Privilege escalation via SSM policy
description: |
  Detects inline IAM policy updates that grant broad AWS Systems Manager permissions. This can enable command
  execution and control actions across managed instances, representing a potential privilege escalation route.
severity: Medium
status: Available
requiredDataConnectors:
  - connectorId: AWS
    dataTypes:
      - AWSCloudTrail
queryFrequency: 1d
queryPeriod: 1d
triggerOperator: gt
triggerThreshold: 0
tactics:
  - PrivilegeEscalation
relevantTechniques:
  - T1098.003
query: |
    AWSCloudTrail
      | where EventName in ("PutUserPolicy","PutRolePolicy","PutGroupPolicy") and isempty(ErrorCode) and isempty(ErrorMessage)
      | extend PolicyName = tostring(parse_json(RequestParameters).policyName)
      | extend Statement = parse_json(tostring((parse_json(RequestParameters).policyDocument))).Statement
      | mvexpand Statement
      | extend Action = parse_json(Statement).Action , Effect = tostring(parse_json(Statement).Effect), Resource = tostring(parse_json(Statement).Resource), Condition = tostring(parse_json(Statement).Condition)
      | extend Action = tostring(Action)
      | where Effect =~ "Allow" and (Action contains "ssm:*" or Action contains "ssm:Create*" or Action contains "ssm:CreateAssociation") and Resource == "*" and isempty(Condition)
      | extend UserIdentityArn = iif(isempty(UserIdentityArn), tostring(parse_json(Resources)[0].ARN), UserIdentityArn)
      | extend UserName = tostring(split(UserIdentityArn, '/')[-1])
      | extend AccountName = case( UserIdentityPrincipalid == "Anonymous", "Anonymous", isempty(UserIdentityUserName), UserName, UserIdentityUserName)
      | extend AccountName = iif(AccountName contains "@", tostring(split(AccountName, '@', 0)[0]), AccountName),
        AccountUPNSuffix = iif(AccountName contains "@", tostring(split(AccountName, '@', 1)[0]), "")
      | distinct TimeGenerated, EventName, PolicyName, SourceIpAddress, RecipientAccountId, AccountName, AccountUPNSuffix, UserIdentityArn, UserIdentityUserName
entityMappings:
  - entityType: Account
    fieldMappings:
      - identifier: Name
        columnName: AccountName
      - identifier: UPNSuffix
        columnName: AccountUPNSuffix
      - identifier: CloudAppAccountId
        columnName: RecipientAccountId
  - entityType: IP
    fieldMappings:
      - identifier: Address
        columnName: SourceIpAddress
customDetails:
  PolicyName: PolicyName
  EventName: EventName
  UserIdentityArn: UserIdentityArn
  RecipientAccountId: RecipientAccountId
alertDetailsOverride:
  alertDisplayNameFormat: 'AWS SSM privilege escalation policy update by {{AccountName}}'
  alertDescriptionFormat: 'Detected {{EventName}} Event, updating inline SSM escalation policy {{PolicyName}} in account {{RecipientAccountId}}.'
version: 1.0.2
kind: Scheduled

Stages and Predicates

Stage 1: source

AWSCloudTrail

Stage 2: where

| where EventName in ("PutUserPolicy","PutRolePolicy","PutGroupPolicy") and isempty(ErrorCode) and isempty(ErrorMessage)

Stage 3: extend

| extend PolicyName = tostring(parse_json(RequestParameters).policyName)

Stage 4: extend

| extend Statement = parse_json(tostring((parse_json(RequestParameters).policyDocument))).Statement

Stage 5: mv-expand

| mvexpand Statement

Stage 6: extend

| extend Action = parse_json(Statement).Action , Effect = tostring(parse_json(Statement).Effect), Resource = tostring(parse_json(Statement).Resource), Condition = tostring(parse_json(Statement).Condition)

Stage 7: extend

| extend Action = tostring(Action)

Stage 8: where

| where Effect =~ "Allow" and (Action contains "ssm:*" or Action contains "ssm:Create*" or Action contains "ssm:CreateAssociation") and Resource == "*" and isempty(Condition)

Stage 9: extend (4 consecutive steps)

| extend UserIdentityArn = iif(isempty(UserIdentityArn), tostring(parse_json(Resources)[0].ARN), UserIdentityArn)
| extend UserName = tostring(split(UserIdentityArn, '/')[-1])
| extend AccountName = case( UserIdentityPrincipalid == "Anonymous", "Anonymous", isempty(UserIdentityUserName), UserName, UserIdentityUserName)
| extend AccountName = iif(AccountName contains "@", tostring(split(AccountName, '@', 0)[0]), AccountName),
    AccountUPNSuffix = iif(AccountName contains "@", tostring(split(AccountName, '@', 1)[0]), "")

Stage 10: distinct

| distinct TimeGenerated, EventName, PolicyName, SourceIpAddress, RecipientAccountId, AccountName, AccountUPNSuffix, UserIdentityArn, UserIdentityUserName

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.

FieldKindValues
Actioncontains
  • ssm:*
  • ssm:Create*
  • ssm:CreateAssociation
Conditionis_null
  • (no value, null check)
Effecteq
  • Allow
ErrorCodeis_null
  • (no value, null check)
ErrorMessageis_null
  • (no value, null check)
EventNamein
  • PutGroupPolicy transforms: cased
  • PutRolePolicy transforms: cased
  • PutUserPolicy transforms: cased
Resourceeq
  • * 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.

FieldSource
PolicyNameextend
Statementextend
Actionextend
Conditionextend
Effectextend
Resourceextend
UserIdentityArnextend
UserNameextend
AccountNameextend
AccountUPNSuffixextend