Splunk Security
@SplunkSecurity35 followers31,487 downloads1,606 copies11 likes89,425 views
2,508 detections
Filters
Last updated
All Time
Detection languages
2,508
Categories
836
532
264
239
208
Platforms
1,611
262
203
142
100
Products / Services
1,231
900
795
217
212
MITRE Techniques
932
664
512
499
306
CVEs
18
13
12
10
9
Detects data leakage and exfiltration attempts targeting Ollama model metadata and configuration endpoints. Adversaries repeatedly query /api/show, /api/tags, and /api/v1/models to systematically extract sensitive model information including architecture details, fine-tuning parameters, system paths, Modelfile configurations, and proprietary customizations. Multiple inspection attempts within a 15-minute window indicate automated exfiltration of valuable intellectual property such as custom model configurations, system prompts, and internal model specifications. This activity represents unauthorized data disclosure that could enable competitive intelligence gathering, model replication, or preparation for advanced attacks against the AI infrastructure.
The analytic detects the creation of image files (PNG, JPG, BMP, GIF, TIFF) in the AppData\Roaming folder by processes that also have a file reference in the same folder. This behavior is commonly associated with malware, such as the Remcos RAT, which captures screenshots and stores them in the AppData folder before exfiltrating them to a command-and-control server. If confirmed malicious, this activity could indicate unauthorized data capture and exfiltration, compromising sensitive information and user privacy.
The following analytic detects the creation of .iso.lnk files in the %USER%\AppData\Local\Temp\<random folder name>\ path, indicating that an ISO file has been mounted and accessed. This detection leverages the Endpoint.Filesystem data model, specifically monitoring file creation events in the Windows Recent folder. This activity is significant as it may indicate the delivery and execution of potentially malicious payloads via ISO files. If confirmed malicious, this could lead to unauthorized code execution, data exfiltration, or further system compromise.
The analytic detects Microsoft OneNote spawning `mshta.exe`, a behavior often associated with spearphishing attacks. This activity is significant as it is commonly used by malware families like TA551, AsyncRat, Redline, and DCRAT to execute malicious scripts. If confirmed malicious, this could allow attackers to execute arbitrary code, potentially leading to data exfiltration, system compromise, or further malware deployment. Immediate investigation and containment are recommended.
Detects abnormal memory allocation patterns and excessive runner operations in Ollama that may indicate resource exhaustion attacks, memory abuse through malicious model loading, or attempts to degrade system performance by overwhelming GPU/CPU resources. Adversaries may deliberately load multiple large models, trigger repeated model initialization cycles, or exploit memory allocation mechanisms to exhaust available system resources, causing denial of service conditions or degrading performance for legitimate users.
The following analytic detects the execution of hh.exe (HTML Help) using InfoTech Storage Handlers to load Windows script code from a Compiled HTML Help (CHM) file. This detection leverages data from Endpoint Detection and Response (EDR) agents, focusing on process names and command-line executions. This activity is significant because it can be used to execute malicious scripts embedded within CHM files, potentially leading to code execution. If confirmed malicious, this technique could allow an attacker to execute arbitrary code, escalate privileges, or persist within the environment.
The following analytic detects the loading of drivers from suspicious paths, which is a technique often used by malicious software such as coin miners (e.g., xmrig). It leverages Sysmon EventCode 6 to identify drivers loaded from non-standard directories. This activity is significant because legitimate drivers typically reside in specific system directories, and deviations may indicate malicious activity. If confirmed malicious, this could allow an attacker to execute code at the kernel level, potentially leading to privilege escalation, persistence, or further system compromise.
This analytic detects the creation of specific archived files, 'passff.tar' and 'cookie.tar', which are known indicators of IcedID malware exfiltrating sensitive browser information like history and cookies. The detection leverages Sysmon EventCode 11 to identify these file creations. This activity is critical as it signifies the potential compromise of sensitive user data, which could lead to further exploitation, data breaches, or escalation of attacker presence within the network.
The following analytic detects the creation, modification, or enabling of scheduled tasks with known suspicious or malicious task names. It leverages Windows Security EventCode 4698, 4700, and 4702 to identify when such tasks are registered, modified, or enabled. This activity is significant as it may indicate an attempt to establish persistence or execute malicious commands on a system. If confirmed malicious, this could allow an attacker to maintain access, execute arbitrary code, or escalate privileges, posing a severe threat to the environment.
Detect when a user creates a new Discretionary Access Control List (DACL) in Active Directory for their own AD object. This can be an indicator of an adversary attempting to maintain persistence or escalate privileges by modifying permissions on their own account to grant themselves additional rights or to prevent other users from modifying their account.
Page 77 of 251