Azure Machine Learning Write Operations

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

Shows the most prevalent users who perform write operations on Azure Machine Learning resources. List the common source IP address for each of those accounts. If an operation is not from those IP addresses, it may be worthy of investigation.

MITRE ATT&CK coverage

Tactic	Techniques
Initial Access	`T1078` Valid Accounts
Execution	`T1059` Command and Scripting Interpreter
Impact	`T1496` Resource Hijacking

Event coverage

Provider	Event	Title
Azure-Microsoft.MachineLearningServices	_catch_all	Azure Machine Learning (catch-all)

Rule body kusto

id: 68c89998-8052-4c80-a1f6-9d81060b6d57
name: Azure Machine Learning Write Operations
description: |
  'Shows the most prevalent users who perform write operations on Azure Machine Learning resources. List the common source IP address for each of those accounts. If an operation is not from those IP addresses, it may be worthy of investigation.'
severity: Low
status: Available
requiredDataConnectors:
  - connectorId: AzureActivity
    dataTypes:
      - AzureActivity
tactics:
  - InitialAccess
  - Execution
  - Impact
relevantTechniques:
  - T1078
  - T1059
  - T1496
queryFrequency: 1d  # Added queryFrequency property
queryPeriod: 1d
triggerOperator: gt
triggerThreshold: 0
query: |
  AzureActivity
  | where ResourceProviderValue == "MICROSOFT.MACHINELEARNINGSERVICES"  // Filter activities related to Microsoft Machine Learning Services
  | extend SCOPE = tostring(parse_json(Authorization).scope)  // Parse Authorization scope as string
  | extend subname = split(Hierarchy, "/")  // Split Hierarchy to extract Subscription Name and ID
  | extend ['Subscription Name'] = subname[-2], ['Subscription ID'] = subname[-1]  // Extract Subscription Name and ID
  | extend Properties = parse_json(Properties)  // Parse Properties as JSON
  | extend Properties_entity = tostring(Properties.entity)  // Cast Properties.entity to string
  | where isnotempty(Properties_entity)  // Filter activities where Properties.entity is not empty
  // | where Properties_entity contains "deepseek"  // Filter activities where Properties.entity contains "deepseek"
  | where OperationNameValue contains "write"  // Filter activities where OperationNameValue contains "write"
  | where OperationNameValue !contains "MICROSOFT.AUTHORIZATION/ROLEASSIGNMENTS/WRITE"  // Exclude role assignments
  | extend LLM = tostring(split(Properties_entity, "/")[-1])  // Extract the last segment of Properties_entity and cast it to string
  | distinct TimeGenerated, tostring(['Subscription Name']), ResourceGroup, tostring(['Subscription ID']), Caller, CallerIpAddress, OperationNameValue, LLM, _ResourceId  // Select distinct relevant fields for output
eventGroupingSettings:
  aggregationKind: SingleAlert
entityMappings:
  - entityType: Account
    fieldMappings:
      - identifier: Name
        columnName: Caller
  - entityType: IP
    fieldMappings:
      - identifier: Address
        columnName: CallerIpAddress
  - entityType: AzureResource
    fieldMappings:
      - identifier: ResourceId
        columnName: _ResourceId  
version: 1.0.0
kind: Scheduled

Stages and Predicates

Stage 1: `source`

AzureActivity

Stage 2: `where`

| where ResourceProviderValue == "MICROSOFT.MACHINELEARNINGSERVICES"

Stage 3: `extend` (5 consecutive steps)

| extend SCOPE = tostring(parse_json(Authorization).scope)
| extend subname = split(Hierarchy, "/")
| extend ['Subscription Name'] = subname[-2], ['Subscription ID'] = subname[-1]
| extend Properties = parse_json(Properties)
| extend Properties_entity = tostring(Properties.entity)

Stage 4: `where`

| where isnotempty(Properties_entity)

Stage 5: `where`

| where OperationNameValue contains "write"

Stage 6: `where`

| where OperationNameValue !contains "MICROSOFT.AUTHORIZATION/ROLEASSIGNMENTS/WRITE"

Stage 7: `extend`

| extend LLM = tostring(split(Properties_entity, "/")[-1])

Stage 8: `distinct`

| distinct TimeGenerated, tostring(['Subscription Name']), ResourceGroup, tostring(['Subscription ID']), Caller, CallerIpAddress, OperationNameValue, LLM, _ResourceId

Exclusions

Top-level NOT(...) conjuncts: predicates this rule actively suppresses.

Field	Kind	Excluded values
`OperationNameValue`	contains	`MICROSOFT.AUTHORIZATION/ROLEASSIGNMENTS/WRITE`

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.

Field	Kind	Values
`OperationNameValue`	contains	`write`
`Properties_entity`	is_not_null	(no value, null check)
`ResourceProviderValue`	eq	`MICROSOFT.MACHINELEARNINGSERVICES` 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.

Field	Source
`SCOPE`	`extend`
`subname`	`extend`
`Subscription ID`	`extend`
`Subscription Name`	`extend`
`Properties`	`extend`
`Properties_entity`	`extend`
`LLM`	`extend`

`j` / `k`	Scroll down / up
`d` / `u`	Half-page down / up
`gg` / `G`	Top / bottom
`h` / `l`	History back / forward
`f`	Follow link (`Shift` = new tab)
`/`	Focus search
`?`	Toggle this help
`↑` / `↓`	Navigate search results
`Enter`	Open highlighted result
`Esc`	Close results / dialog

`type:`	`events` / `rules` / `providers`
`vendor:`	`sigma` / `elastic` / `splunk` / `kusto` / `chronicle` (vendor name alone also works: `sigma:`, `kql:`, `secops:`…)
`tactic:`	TA-id, slug, or name: `credential_access`, `TA0006`
`technique:`	technique or sub-technique ID: `T1003`, `T1003.001` (alias `tech:`)
`severity:`	`critical` / `high` / `medium` / `low` / `informational` (alias `sev:`)
`risk_score`	Numeric comparison on the Elastic risk score (0 to 100): `risk_score>50`, `risk_score<=20`, `risk_score=99` (alias `risk`; Elastic rules only)
`stages:`	Rules with exactly N pipeline stages
`correlation:`	`single_event` / `sequence` / `alternatives` / `alternatives_cross_log` / `all_required` / `correlated`
`with:`	Co-occurrence event-id; stacks (`with:4624 with:4769`) to require all, while a comma list in one occurrence (`with:4624,4769`) is an either-or group. Implies multi-event
`like:`	Structural neighbors of a rule slug (equivalents + subsumption stricter / broader): `like:comsvcs_lsass_memory_dump-splunk-sysmon`
`groupby:`	Entity-grouping substring match against `group_by_keys`: `groupby:user`, `groupby:host`
`uses:`	Rules whose predicate tree touches the field (any kind, any value): `uses:CommandLine`
`excludes:`	Rules with top-level `not()` clauses on the field (FP whitelists): `excludes:ParentImage`
`field:` / `value:`	Predicate search; narrows rule cards to those with a matching leaf and drives the indicator tier. Unquoted = substring, wildcards allowed (`value:mimikatz`)
`indicator:`	Shorthand for `field:F value:V`: `indicator:Image=*\powershell.exe`
`kind:`	Filter by predicate kind. Narrows rule cards to those carrying a matching predicate leaf (`vendor:elastic kind:cidr_match`) and drives the indicator tier: `contains` / `starts_with` / `ends_with` / `regex` / `cidr` / `eq` / `in` … (operator aliases `op:`/`match:`)
`has:` / `no:`	`sample`, `field`, `notes`, `refs`, `trace`, `thirdparty`, `rule`, `pattern`, `timewindow`, `threshold`, `newterms`, `sigma`/`elastic`/`splunk`/`kusto`/`chronicle`
`-op:val`	Exclude matches; works on most operators but not `type:`/`like:`/`has:`/`no:` (use `no:<flag>` to exclude a rule flag): `tactic:execution -vendor:splunk`. Standalone `-kind:`/`-field:`/`-value:` drop every rule carrying a matching predicate leaf (`type:rules -kind:is_null`)
`field:"…"` / `value:"…"`	Quoted value = anchored exact match (also allows spaces): `value:"net user"`
`a,b`	Comma = OR inside one operator (`vendor:sigma,elastic`, `severity:high,critical`); repeating a facet merges the same way. `field:`/`value:` never split (literal commas)
`vendors:` / `stage:`	Singular and plural spellings fold to the canonical operator and value: `tactics:` = `tactic:`, `type:event` = `type:events`, `correlation:sequences` = `correlation:sequence`, `has:thresholds` = `has:threshold`
`"quoted phrase"`	Exact-match a multi-word phrase (free text)

Azure Machine Learning Write Operations

MITRE ATT&CK coverage

Event coverage

Rule body kusto

Stages and Predicates

Stage 1: `source`

Stage 2: `where`

Stage 3: `extend` (5 consecutive steps)

Stage 4: `where`

Stage 5: `where`

Stage 6: `where`

Stage 7: `extend`

Stage 8: `distinct`

Exclusions

Indicators

Output fields

Keyboard shortcuts

Search operators

Azure Machine Learning Write Operations

MITRE ATT&CK coverage

Event coverage

Rule body kusto

Stages and Predicates

Stage 1: source

Stage 2: where

Stage 3: extend (5 consecutive steps)

Stage 4: where

Stage 5: where

Stage 6: where

Stage 7: extend

Stage 8: distinct

Exclusions

Indicators

Output fields

Stage 1: `source`

Stage 2: `where`

Stage 3: `extend` (5 consecutive steps)

Stage 4: `where`

Stage 5: `where`

Stage 6: `where`

Stage 7: `extend`

Stage 8: `distinct`