PowerShell Loading DotNET into Memory via Reflection

SPLUNK detection rule by Michael Haag, Teoderick Contreras Splunk · Apache-2.0

The following analytic detects the use of PowerShell scripts to load .NET assemblies into memory via reflection, a technique often used in malicious activities such as those by Empire and Cobalt Strike. It leverages PowerShell Script Block Logging (EventCode=4104) to capture and analyze the full command executed. This behavior is significant as it can indicate advanced attack techniques aiming to execute code in memory, bypassing traditional defenses. If confirmed malicious, this activity could lead to unauthorized code execution, privilege escalation, and persistent access within the environment.

name: PowerShell Loading DotNET into Memory via Reflection
id: 85bc3f30-ca28-11eb-bd21-acde48001122
version: 19
creation_date: '2021-06-15'
modification_date: '2026-06-29'
author: Michael Haag, Teoderick Contreras Splunk
status: production
type: Anomaly
description: |-
    The following analytic detects the use of PowerShell scripts to load .NET assemblies into memory via reflection, a technique often used in malicious activities such as those by Empire and Cobalt Strike.
    It leverages PowerShell Script Block Logging (EventCode=4104) to capture and analyze the full command executed.
    This behavior is significant as it can indicate advanced attack techniques aiming to execute code in memory, bypassing traditional defenses.
    If confirmed malicious, this activity could lead to unauthorized code execution, privilege escalation, and persistent access within the environment.
data_source:
    - Powershell Script Block Logging 4104
search: |-
    `powershell` EventCode=4104
    ScriptBlockText IN (
        "*.LoadFrom(*",
        "*.LoadModule(*",
        "*.LoadWithPartialName*",
        "*Reflection.Assembly.Load*",
        "*Reflection.Assembly]::('daoL'[-1..-4] -join '')*",
        "*Reflection.Assembly]::Load*",
        "*ReflectionOnlyLoad*",
        "*UnsafeLoadFrom*"
    )

    | fillnull
    | stats count min(_time) as firstTime
                  max(_time) as lastTime
      by dest signature signature_id user_id
         vendor_product EventID Guid Opcode
         Name Path ProcessID
         ScriptBlockId ScriptBlockText

    | `security_content_ctime(firstTime)`
    | `security_content_ctime(lastTime)`
    | `powershell_loading_dotnet_into_memory_via_reflection_filter`
how_to_implement: To successfully implement this analytic, you will need to enable PowerShell Script Block Logging on some or all endpoints. Additional setup here https://help.splunk.com/en/security-offerings/splunk-user-behavior-analytics/get-data-in/5.4.1/add-other-data-to-splunk-uba/configure-powershell-logging-to-see-powershell-anomalies-in-splunk-uba.
known_false_positives: False positives should be limited as day to day scripts do not use this method.
references:
    - https://docs.microsoft.com/en-us/dotnet/api/system.reflection.assembly?view=net-5.0
    - https://help.splunk.com/en/security-offerings/splunk-user-behavior-analytics/get-data-in/5.4.1/add-other-data-to-splunk-uba/configure-powershell-logging-to-see-powershell-anomalies-in-splunk-uba.
    - https://blog.palantir.com/tampering-with-windows-event-tracing-background-offense-and-defense-4be7ac62ac63
    - https://static1.squarespace.com/static/552092d5e4b0661088167e5c/t/59c1814829f18782e24f1fe2/1505853768977/Windows+PowerShell+Logging+Cheat+Sheet+ver+Sept+2017+v2.1.pdf
    - https://www.crowdstrike.com/blog/investigating-powershell-command-and-script-logging/
drilldown_searches:
    - name: View the detection results for - "$dest$" and "$user_id$"
      search: '%original_detection_search% | search  dest = "$dest$" user_id = "$user_id$"'
      earliest_offset: $info_min_time$
      latest_offset: $info_max_time$
    - name: View risk events for the last 7 days for - "$dest$" and "$user_id$"
      search: '| from datamodel Risk.All_Risk | search normalized_risk_object IN ("$$", "$user_id$") | stats count min(_time) as firstTime max(_time) as lastTime values(search_name) as "Search Name" values(risk_message) as "Risk Message" values(analyticstories) as "Analytic Stories" values(annotations._all) as "Annotations" values(annotations.mitre_attack.mitre_tactic) as "ATT&CK Tactics" by normalized_risk_object | `security_content_ctime(firstTime)` | `security_content_ctime(lastTime)`'
      earliest_offset: 7d
      latest_offset: '0'
intermediate_findings:
    entities:
        - field: dest
          type: system
          score: 20
          message: A suspicious powershell script with the ScriptBlockId [$ScriptBlockId$] containing reflective class assembly commands to load .net code in memory was executed on host $dest$
        - field: user_id
          type: user
          score: 20
          message: A suspicious powershell script with the ScriptBlockId [$ScriptBlockId$] containing reflective class assembly commands to load .net code in memory was executed on host $dest$ by user $user_id$
analytic_story:
    - Winter Vivern
    - AgentTesla
    - AsyncRAT
    - Hermetic Wiper
    - Malicious PowerShell
    - Data Destruction
    - 0bj3ctivity Stealer
    - Hellcat Ransomware
    - Axios Supply Chain Post Compromise
    - VIP Keylogger
asset_type: Endpoint
mitre_attack_id:
    - T1059.001
product:
    - Splunk Enterprise
    - Splunk Enterprise Security
    - Splunk Cloud
category: endpoint
security_domain: endpoint
tests:
    - name: True Positive Test
      attack_data:
        - data: https://media.githubusercontent.com/media/splunk/attack_data/master/datasets/attack_techniques/T1059.001/powershell_script_block_logging/reflection.log
          source: XmlWinEventLog:Microsoft-Windows-PowerShell/Operational
          sourcetype: XmlWinEventLog
      test_type: unit

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

788ca6b1-a3d3-425b-87c7-e1a26ef5cfe3

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