Detections
Explore public detection logic contributed by the community across SIEM and rule languages.
1,695 detections
Filters
Last updated
All Time
Detection languages
547
535
293
77
61
Contributors
401
223
219
189
154
Categories
727
561
461
367
323
Platforms
1,477
157
136
87
8
Products / Services
335
331
222
196
56
MITRE Techniques
18,127
15,648
12,596
8,407
6,231
1,695
CVEs
3
1
1
1
1
IDS Classtypes
26
10
7
6
2
IDS Protocols
42
11
Detects the execution of Python interpreters with command-line arguments that utilize common obfuscation and decoding techniques. The rule identifies a combination of Base64 decoding, codec decoding, ROT13/ROT47 rotation, and hexadecimal string conversion, which are frequently used to hide malicious scripts or payloads from static analysis.
Detects anomalous HTTP responses containing high-entropy, potentially multi-layered obfuscated payloads. The rule specifically monitors for patterns suggesting Base64, ROT13, and Hex encoding in the response body, which is consistent with methods used to deliver obfuscated malicious AI skills or scripts via web traffic.
Detects instances where node.exe or npm.exe processes spawn PowerShell to execute download-related cmdlets (e.g., Invoke-WebRequest, IEX) or encoded commands, followed by the execution of a process from user-writable directories (TEMP or AppData) within a one-minute window, a pattern associated with PostCSS typosquatting supply chain attacks.
Detects the execution of Python interpreters with command-line arguments that utilize common obfuscation and decoding techniques. The rule identifies a combination of Base64 decoding, codec decoding, ROT13/ROT47 rotation, and hexadecimal string conversion, which are frequently used to hide malicious scripts or payloads from static analysis.
Detects anomalous HTTP responses containing high-entropy, potentially multi-layered obfuscated payloads. The rule specifically monitors for patterns suggesting Base64, ROT13, and Hex encoding in the response body, which is consistent with methods used to deliver obfuscated malicious AI skills or scripts via web traffic.
Detects modern phishing tradecraft where malicious JavaScript is embedded inside SVG image files. The browser evaluates the SVG, decodes a Base64 payload, and drops a secondary archive or executable directly to disk via Blob/Data URIs. This rule specifically looks for file creation events in the Downloads folder for common executable or archive file types, correlated with browser processes (Chrome, Edge, Firefox) accessing local SVG files.
Detects the creation of potentially malicious archive files (.zip, .iso, .img, .cab) in the Downloads folder, initiated by web browsers (Chrome, Edge, Firefox), where the initiating process's command line indicates a local file access (file:///), suggesting a Blob-to-Disk dropper technique often associated with AI-obfuscated phishing campaigns.
Detects execution patterns associated with the Miasma/Mini Shai-Hulud multi-layer obfuscated npm payload. This includes monitoring for bun runtime spawning from Node/NPM contexts, execution of bun with specific patterns from /tmp, downloading of specific bun versions from GitHub, and execution of suspicious index.js scripts within .claude directories.
Detects the creation of executable files within the C:\ProgramData\ directory initiated by common command-line interpreters (cmd.exe, powershell.exe, wscript.exe, svchost.exe). This pattern is often indicative of the SystemBC backdoor or similar malware attempting to drop its executable payload in a non-standard, persistent location to evade initial detection.
Detects a malicious post-install technique where an npm package utilizes node.exe to perform an 'npm pack' and overwrite critical files within the node_modules directory. This behavior, observed in the Miasma/Polymarket Trap, is designed to bypass static analysis performed during initial installation by swapping legitimate files with malicious variants after the fact.
Detects the final payload stage of LokiBot infostealer, characterized by a 32-bit Windows PE file built with MSVC 2015. The rule identifies specific known malicious file hashes or a structural signature featuring a non-standard PE section named 'x', which is used by the malware to store 3DES-encrypted C2 endpoint addresses.
Detects the LokiBot malware attempting to resolve the 'CreateMutexW' API function at runtime using a specific hashed value (0xcf167df4). This technique is used by the malware to prevent multiple concurrent instances from running on the same host, which is a characteristic behavioral pattern for this information stealer.
Detects PowerShell scripts that contain decryption routines and in-memory execution commands, a technique used by UNC5142 to run final payloads while evading disk-based detection.
Detects the use of certutil.exe with command-line arguments indicative of downloading files from a URL or decoding/encoding data. This utility is often abused by adversaries for ingress tool transfer or obfuscation.
Detects a game-related process creating ZIP, 7z, or encrypted data containers in an inconspicuous AppData subdirectory. This behavior is indicative of data staging prior to exfiltration, often seen in supply chain attacks or targeted espionage.
Detects the creation of scheduled tasks using 'schtasks.exe' with command line arguments that indicate suspicious activity. This includes tasks created in common temporary or user-writable directories, or tasks that involve PowerShell with encoded commands or obfuscated strings, which are often used by adversaries for persistence or execution.
Detects HTTP responses containing JavaScript files with a Content-Length exceeding 1MB. This behavior is associated with the Bluekit PhaaS (Phishing-as-a-Service) platform, which uses large, obfuscated JavaScript bundles for browser fingerprinting and to evade static security signatures.
Detects the execution of malicious .NET assemblies associated with the Turla group's STOCKSTAY malware family. The rule identifies specific file names (ClientMNGR2.exe, GR3.exe) and known SHA256 hashes indicative of these assemblies, which utilize the K1MORPHER string obfuscation technique.
Identifies processes executing Base64 decoded commands where the parent script recently communicated with an AI endpoint, representing prompt-injected or AI-generated payloads.
Detects a suspicious PowerShell command sequence involving network downloads to temporary directories followed by the execution of a Node.js binary from within the user's AppData directory within a 30-minute window. This behavior is indicative of a dropper pattern where a malicious script retrieves and executes a secondary stage payload.
Detects a behavioral pattern associated with LegionLoader malware, where a process running from the 'AppData\Roaming' directory loads 'user32.dll' or 'kernel32.dll'. This pattern is frequently used to dynamically resolve GUI interaction APIs (such as MessageBoxA) via GetProcAddress as an anti-analysis and obfuscation technique.
Page 48 of 81