Skip to content

Siemphony is in beta and still being built. How a rule earns its badge.

Siemphony’s repertoire

Per-user CLSID registered with a payload under InprocServer32 or LocalServer32

AN1323's registry source is Security EventID 4657, but this rule is written in Sysmon's registry_set vocabulary (a single TargetObject) because that is what the brief's sigmaLogsource declares; a deployment feeding from 4657 instead needs its ObjectName/ObjectValueName fields mapped onto TargetObject first — enabling Audit Registry and a SACL alone does not make 4657 fit this shape. The match targets the actual hijack mechanism — a CLSID under `HKEY_CURRENT_USER\Software\Classes\CLSID\` (Sysmon's `HKU\` prefix) overriding the machine-wide HKLM registration — and ends on the CLSID's InprocServer32/LocalServer32 subkey and its `(Default)` value, because that is where the payload path actually lives; the CLSID key's own value carries nothing. The module load or execution MITRE correlates against this write is a cross-event join Sigma cannot express, and most public Sysmon configs filter `\Software\Classes` out as noise, so this needs an explicit include rule first. UNVERIFIED AS AUTHORED — derived from MITRE ATT&CK DET0481, not hand-written and not tested by its author. Any lab result is on this rule's own page.Full description

The detection

The 4 SIEM queries below are previews produced by Siemphony’s own translator, not pySigma, and none has been executed against a real backend — review before you deploy.

detection.yml

title: Per-user CLSID registered with a payload under InprocServer32 or LocalServer32id: c9951408-0230-4b5c-8e36-0bcd444ed995status: experimentaldescription: |  AN1323's registry source is Security EventID 4657, but this rule is written  in Sysmon's registry_set vocabulary (a single TargetObject) because that is  what the brief's sigmaLogsource declares; a deployment feeding from 4657  instead needs its ObjectName/ObjectValueName fields mapped onto TargetObject  first — enabling Audit Registry and a SACL alone does not make 4657 fit this  shape. The match targets the actual hijack mechanism — a CLSID under  `HKEY_CURRENT_USER\Software\Classes\CLSID\` (Sysmon's `HKU\` prefix)  overriding the machine-wide HKLM registration — and ends on the CLSID's  InprocServer32/LocalServer32 subkey and its `(Default)` value, because that  is where the payload path actually lives; the CLSID key's own value carries  nothing. The module load or execution MITRE correlates against this write is  a cross-event join Sigma cannot express, and most public Sysmon configs  filter `\Software\Classes` out as noise, so this needs an explicit include  rule first.  UNVERIFIED AS AUTHORED — derived from MITRE ATT&CK DET0481, not hand-written and not tested by its author. Any lab result is on this rule's own page.references:  - https://attack.mitre.org/techniques/T1546/015  - https://attack.mitre.org/detectionstrategies/DET0481author: Siemphony pipeline (machine-derived from MITRE ATT&CK v19.2)date: 2026-08-16tags:  - attack.privilege-escalation  - attack.persistence  - attack.t1546.015logsource:  category: registry_set  product: windowsdetection:  selection_hive:    TargetObject|startswith: 'HKU\'  selection_com_path:    TargetObject|contains: '\Software\Classes\CLSID\'  selection_payload_value:    TargetObject|endswith:      - '\InprocServer32\(Default)'      - '\LocalServer32\(Default)'  condition: selection_hive and selection_com_path and selection_payload_valuefalsepositives:  - "Application installers and updaters that register per-user COM components without needing admin rights — browser toolbars and extensions, Office and Adobe shell-integration add-ins, and cloud-storage overlay-icon handlers all write InprocServer32 under HKCU as their normal install step, and this is the highest-volume source on a general-purpose fleet."  - "Non-persistent VDI, App-V and roaming-profile environments that reprovision per-user COM registrations from a template on every logon, producing a recurring hit for the same CLSID across every session on every host."  - "Shell-extension and productivity tools that deliberately use a per-user CLSID to avoid requiring elevation — 7-Zip, WinRAR and similar archivers, and cloud-sync clients placing a Explorer overlay handler, are legitimate uses of the exact registration shape this rule is written to catch."  - "Endpoint security and browser-monitoring agents that register their own per-user COM hooks for legitimate inspection, which are structurally identical to a hijack from this rule's point of view."level: medium

Sentinel · KQL

Run this as a search.

DeviceRegistryEvents| where ((RegistryKey startswith "HKU\\" and RegistryKey contains "\\Software\\Classes\\CLSID\\") and (RegistryKey endswith "\\InprocServer32\\(Default)" or RegistryKey endswith "\\LocalServer32\\(Default)"))

Splunk · SPL

Run this as a search.

index=* ((TargetObject="HKU\\*" AND TargetObject="*\\Software\\Classes\\CLSID\\*") AND (TargetObject="*\\InprocServer32\\(Default)" OR TargetObject="*\\LocalServer32\\(Default)"))

Elastic · ES|QL

Run this as a search.

FROM logs-*| WHERE ((TO_LOWER(registry.path) LIKE "hku\\\\*" AND TO_LOWER(registry.path) LIKE "*\\\\software\\\\classes\\\\clsid\\\\*") AND (TO_LOWER(registry.path) LIKE "*\\\\inprocserver32\\\\(default)" OR TO_LOWER(registry.path) LIKE "*\\\\localserver32\\\\(default)"))

Wazuh · XML rule

Deploy to your manager — this is a rule, not a search.

<group name="sigma,windows,registry_set,">  <!-- Rule ids must be unique on your manager; 100000+ is the user range. -->  <rule id="100000" level="7">    <!-- Set <if_sid> to the decoder/base rule for windows so this only evaluates relevant events. -->    <field name="TargetObject" type="pcre2">(?i)^HKU\\</field>    <field name="TargetObject" type="pcre2">(?i)\\Software\\Classes\\CLSID\\</field>    <field name="TargetObject" type="pcre2">(?i)(\\InprocServer32\\\(Default\)$|\\LocalServer32\\\(Default\)$)</field>    <description>Per-user CLSID registered with a payload under InprocServer32 or LocalServer32</description>    <mitre>      <id>T1546.015</id>    </mitre>  </rule></group>

Verdicts · reactions · comments

Community

Verdicts from engineers who actually deployed it, and the conversation around it.

Log in to react
Not yet reported on

Nobody has run this in a real environment and said what happened.

Verdicts from engineers who deployed it

0 cast

No verdicts reported yet. The first one is worth more than the tenth — say what you ran it against.

Log in to report a verdict.

Log in to join the discussion.

No comments yet. Someone who deploys this will have something to say about it.