LSASS Process Access via Windows API

  1[metadata]
  2creation_date = "2023/03/02"
  3integration = ["endpoint", "m365_defender"]
  4maturity = "production"
  5min_stack_comments = "MV_CONTAINS was added to ES|QL in 9.2.0 and is not available on 8.19; 9.3.0 is the lowest supported 9.x release."
  6min_stack_version = "9.3.0"
  7updated_date = "2026/09/22"
  8[transform]
  9[[transform.osquery]]
 10label = "Osquery - Retrieve DNS Cache"
 11query = "SELECT * FROM dns_cache"
 12
 13[[transform.osquery]]
 14label = "Osquery - Retrieve All Services"
 15query = "SELECT description, display_name, name, path, pid, service_type, start_type, status, user_account FROM services"
 16
 17[[transform.osquery]]
 18label = "Osquery - Retrieve Services Running on User Accounts"
 19query = """
 20SELECT description, display_name, name, path, pid, service_type, start_type, status, user_account FROM services WHERE
 21NOT (user_account LIKE '%LocalSystem' OR user_account LIKE '%LocalService' OR user_account LIKE '%NetworkService' OR
 22user_account == null)
 23"""
 24
 25[[transform.osquery]]
 26label = "Osquery - Retrieve Service Unsigned Executables with Virustotal Link"
 27query = """
 28SELECT concat('https://www.virustotal.com/gui/file/', sha1) AS VtLink, name, description, start_type, status, pid,
 29services.path FROM services JOIN authenticode ON services.path = authenticode.path OR services.module_path =
 30authenticode.path JOIN hash ON services.path = hash.path WHERE authenticode.result != 'trusted'
 31"""
 32
 33
 34[rule]
 35author = ["Elastic"]
 36description = "Identifies access attempts to the LSASS handle, which may indicate an attempt to dump credentials from LSASS memory."
 37from = "now-30m"
 38interval = "15m"
 39language = "esql"
 40license = "Elastic License v2"
 41name = "LSASS Process Access via Windows API"
 42note = """## Triage and analysis
 43
 44### Investigating LSASS Process Access via Windows API
 45
 46The Local Security Authority Subsystem Service (LSASS) is a critical Windows component responsible for managing user authentication and security policies. Adversaries may attempt to access the LSASS handle to dump credentials from its memory, which can be used for lateral movement and privilege escalation.
 47
 48This rule identifies attempts to access LSASS by monitoring for specific API calls (OpenProcess, OpenThread) targeting the "lsass.exe" process.
 49
 50> **Note**:
 51> This investigation guide uses the [Osquery Markdown Plugin](https://www.elastic.co/guide/en/security/current/invest-guide-run-osquery.html) introduced in Elastic Stack version 8.5.0. Older Elastic Stack versions will display unrendered Markdown in this guide.
 52
 53### Possible investigation steps
 54
 55- Investigate other alerts associated with the user/host during the past 48 hours.
 56- Investigate the process execution chain (parent process tree) of the process that accessed the LSASS handle.
 57  - Examine their executable files for prevalence, whether they are located in expected locations, and if they are signed with valid digital signatures.
 58  - Determine the first time the process executable was seen in the environment and if this behavior happened in the past.
 59  - Validate the activity is not related to planned patches, updates, network administrator activity, or legitimate software installations.
 60  - Investigate any abnormal behavior by the subject process, such as network connections, DLLs loaded, registry or file modifications, and any spawned child processes.
 61- Assess the access rights (`process.Ext.api.parameters.desired_access`field) requested by the process. This [Microsoft documentation](https://learn.microsoft.com/en-us/windows/win32/procthread/process-security-and-access-rights) may be useful to help the interpretation.
 62- If there are traces of LSASS memory being successfully dumped, investigate potentially compromised accounts. Analysts can do this by searching for login events (e.g., 4624) to the target host.
 63- Examine the host for derived artifacts that indicate suspicious activities:
 64  - Analyze the executables of the processes using a private sandboxed analysis system.
 65  - Observe and collect information about the following activities in both the sandbox and the alert subject host:
 66    - Attempts to contact external domains and addresses.
 67      - Use the Elastic Defend network events to determine domains and addresses contacted by the subject process by filtering by the process's `process.entity_id`.
 68      - Examine the DNS cache for suspicious or anomalous entries.
 69        - $osquery_0
 70    - Use the Elastic Defend registry events to examine registry keys accessed, modified, or created by the related processes in the process tree.
 71    - Examine the host services for suspicious or anomalous entries.
 72      - $osquery_1
 73      - $osquery_2
 74      - $osquery_3
 75  - Retrieve the files' SHA-256 hash values using the PowerShell `Get-FileHash` cmdlet and search for the existence and reputation of the hashes in resources like VirusTotal, Hybrid-Analysis, CISCO Talos, Any.run, etc.
 76
 77
 78### False positive analysis
 79
 80- If this rule is noisy in your environment due to expected activity, consider adding exceptions — preferably with a combination of `process.executable`, `process.code_signature.subject_name` and `process.Ext.api.parameters.desired_access_numeric` conditions.
 81
 82### Related Rules
 83
 84- Suspicious Lsass Process Access - 128468bf-cab1-4637-99ea-fdf3780a4609
 85- Potential Credential Access via DuplicateHandle in LSASS - 02a4576a-7480-4284-9327-548a806b5e48
 86- LSASS Memory Dump Handle Access - 208dbe77-01ed-4954-8d44-1e5751cb20de
 87
 88### Response and Remediation
 89
 90- Initiate the incident response process based on the outcome of the triage.
 91  - If malicious activity is confirmed, perform a broader investigation to identify the scope of the compromise and determine the appropriate remediation steps.
 92- Isolate the involved host to prevent further post-compromise behavior.
 93- If the triage identified malware, search the environment for additional compromised hosts.
 94  - Implement temporary network rules, procedures, and segmentation to contain the malware.
 95  - Stop suspicious processes.
 96  - Immediately block the identified indicators of compromise (IoCs).
 97  - Inspect the affected systems for additional malware backdoors like reverse shells, reverse proxies, or droppers that attackers could use to reinfect the system.
 98- Remove and block malicious artifacts identified during triage.
 99- Investigate credential exposure on systems compromised or used by the attacker to ensure all compromised accounts are identified. Reset passwords for these accounts and other potentially compromised credentials, such as email, business systems, and web services.
100- Run a full antimalware scan. This may reveal additional artifacts left in the system, persistence mechanisms, and malware components.
101- Reimage the host operating system or restore the compromised files to clean versions.
102- Determine the initial vector abused by the attacker and take action to prevent reinfection through the same vector.
103- Using the incident response data, update logging and audit policies to improve the mean time to detect (MTTD) and the mean time to respond (MTTR).
104"""
105
106setup = """## Setup
107
108This rule is designed for data generated by [Elastic Defend](https://www.elastic.co/security/endpoint-security), which provides native endpoint detection and response, along with event enrichments designed to work with our detection rules.
109
110Setup instructions: https://ela.st/install-elastic-defend
111
112### Additional data sources
113
114This rule also supports the following third-party data sources. For setup instructions, refer to the links below:
115
116- [Microsoft Defender XDR](https://ela.st/m365-defender)
117"""
118
119references = ["https://github.com/redcanaryco/atomic-red-team/blob/master/atomics/T1003.001/T1003.001.md"]
120risk_score = 47
121rule_id = "ff4599cb-409f-4910-a239-52e4e6f532ff"
122severity = "medium"
123tags = [
124    "Domain: Endpoint",
125    "OS: Windows",
126    "Use Case: Threat Detection",
127    "Tactic: Credential Access",
128    "Tactic: Execution",
129    "Data Source: Elastic Defend",
130    "Data Source: Microsoft Defender XDR",
131    "Resources: Investigation Guide",
132    "Noise: Medium",
133    "Performance: Normal",
134    "Rule Type: ES|QL",
135    "Platform: Windows",
136    "Resources: Osquery",
137]
138timestamp_override = "event.ingested"
139type = "esql"
140
141query = '''
142from logs-endpoint.events.api-*, logs-m365_defender.event-* metadata _id, _version, _index
143
144| where mv_contains(event.category, "api") and host.os.type == "windows" and
145   process.Ext.api.name in ("OpenProcess", "OpenThread", "ReadProcessMemory") and
146   Target.process.name == "lsass.exe" and process.executable is not null and
147
148   // Noisy patterns
149   not to_lower(process.executable) like """c:\\program files\\*.exe""" and
150   not to_lower(process.executable) like """c:\\program files (x86)\\*.exe""" and
151   not to_lower(process.executable) like """c:\\programdata\\microsoft\\windows defender\\platform\\msmpeng.exe""" and
152   not to_lower(process.executable) like """c:\\programdata\\microsoft\\windows defender\\platform\\*\\msmpeng.exe"""
153
154 /* normalize process paths to reduce known random patterns in process.executable */
155| eval Esql.process_path = replace(process.executable, """([0-9a-fA-F]{8}-[0-9a-fA-F]{4}-[0-9a-fA-F]{4}-[0-9a-fA-F]{4}-[0-9a-fA-F]{12}|ns[a-z][A-Z0-9]{3,4}\.tmp|DX[A-Z0-9]{3,4}\.tmp|7z[A-Z0-9]{3,5}\.tmp|[0-9\.\-\_]{3,})""", "")
156
157// Group by process path
158| stats Esql.access_count = count(*),
159        Esql.count_distinct_hosts = count_distinct(host.id),
160        Esql.host_id_values = VALUES(host.id),
161        Esql.host_name_values = VALUES(host.name),
162        Esql.user_name_values = VALUES(user.name),
163        Esql.process_pid_values = VALUES(process.entity_id),
164        Esql.process_executable_values = VALUES(process.executable),
165        Esql.data_stream_namespace.values = VALUES(data_stream.namespace),
166        Esql.user_name_values = VALUES(user.name) by Esql.process_path
167
168// Limit to rare instances limited to 1 unique host
169| where Esql.count_distinct_hosts == 1 and Esql.access_count <= 3
170
171// Extract the single host ID and process into their corresponding ECS fields for alerts exclusion
172| eval host.id = mv_min(Esql.host_id_values),
173       host.name = mv_min(Esql.host_name_values),
174       process.executable = mv_min(Esql.process_executable_values), 
175       user.name = mv_min(Esql.user_name_values)
176
177// Add the new field to the keep statement
178| keep Esql.*, host.id, host.name, user.name, process.executable
179'''
180
181
182[[rule.threat]]
183framework = "MITRE ATT&CK"
184[[rule.threat.technique]]
185id = "T1003"
186name = "OS Credential Dumping"
187reference = "https://attack.mitre.org/techniques/T1003/"
188[[rule.threat.technique.subtechnique]]
189id = "T1003.001"
190name = "LSASS Memory"
191reference = "https://attack.mitre.org/techniques/T1003/001/"
192
193
194
195[rule.threat.tactic]
196id = "TA0006"
197name = "Credential Access"
198reference = "https://attack.mitre.org/tactics/TA0006/"
199[[rule.threat]]
200framework = "MITRE ATT&CK"
201[[rule.threat.technique]]
202id = "T1106"
203name = "Native API"
204reference = "https://attack.mitre.org/techniques/T1106/"
205
206
207[rule.threat.tactic]
208id = "TA0002"
209name = "Execution"
210reference = "https://attack.mitre.org/tactics/TA0002/"