Detections
Explore public detection logic contributed by the community across SIEM and rule languages.
1,937 detections
Filters
Last updated
All Time
Detection languages
1,930
3
2
1
1
Contributors
592
477
360
250
124
Categories
995
927
834
546
538
Platforms
1,753
214
81
54
41
Products / Services
1,868
1,144
65
51
17
MITRE Techniques
564
412
305
201
175
CVEs
4
4
4
2
2
IDS Classtypes
1,937
503
461
381
236
IDS Protocols
1,098
362
201
171
37
This rule detects when a file containing both ".plist" in its HTTP URI and "LaunchDaemon" in its content is delivered via HTTP. This combination suggests the download of a macOS LaunchDaemon property list file, which could be used for persistence or privilege escalation on macOS systems.
This rule detects the download of a potential trojanized installer file over HTTP. It looks for HTTP responses to clients that contain 'setup', 'install', or 'installer' in the URI, have a file size greater than 500KB, and contain the 'MZ' header, indicating a Windows executable. This combination suggests a user is downloading an executable that could be malicious.
This rule detects the download of archive files (ZIP, RAR, 7Z, TAR.GZ) over HTTP that contain an executable file signature ('MZ' header). This combination suggests a potential user execution scenario where a malicious executable might be delivered within an archive.
This rule detects when an executable file (.exe) is copied to a share associated with a removable drive (indicated by 'USBSTOR') over the SMB protocol. This could indicate an attempt to transfer malicious executables to or from a removable device.
This rule detects attempts to download typosquatted NPM packages from registry.npmjs.org. It specifically looks for HTTP requests to the NPM registry where the URI contains common typos of popular packages like 'lodash', 'requests', 'express', 'mongoose', and 'momentjs'. This indicates a potential supply chain attack where developers might accidentally download malicious packages due to a typo.
This rule detects HTTP responses containing JavaScript that attempts to redirect the user's browser to another page and executes code using 'eval()'. This combination is often used in malicious web pages to redirect users to exploit kits or phishing sites, and to obfuscate malicious JavaScript.
This rule detects HTTP responses from a web server to a client that contain JavaScript code attempting to redirect the user's browser. Specifically, it looks for the string "window.location.replace" within the HTTP response body (file.data) when the HTTP referer header indicates the request originated from Google or Bing search results. This pattern is often associated with content injection attacks where malicious code is injected into legitimate search results to redirect users to malicious sites.
This rule detects a drive-by download attempt initiated via a malicious advertisement. It specifically looks for HTTP responses to clients where the HTTP referer header contains common advertisement-related keywords (ad., ads., advert) and the response body contains the 'MZ' magic bytes, indicating an executable file download. This combination suggests a user was redirected from an advertisement to a site that then attempted to download an executable.
This rule detects Fast Flux DNS activity, characterized by rapid changes in IP addresses associated with the same domain within a short period. Specifically, it triggers an alert if 30 or more distinct IP addresses are observed for a single domain within a 10-second window, based on DNS queries. This behavior is often indicative of malware using fast flux techniques to evade detection and make takedowns more difficult.
This rule detects potential command and control (C2) communication originating from a compromised serverless function. It specifically looks for HTTP POST requests with a body size greater than 500 bytes, targeting domains associated with serverless platforms like Vercel (.vercel.app), Netlify (.netlify.app), and Cloudflare Pages (.pages.dev). A threshold of 10 such requests within 60 seconds from the same source IP triggers the alert, indicating a possible high-volume data exfiltration or C2 activity.
This rule detects HTTP PUT requests to known remote file hosting services (transfer.sh, file.io, gofile.io, anonfiles). This activity is indicative of payload staging, where an attacker uploads malicious files or tools to a remote server for later use in an attack, often as part of the resource development phase of the attack lifecycle.
This rule detects attempts to upload executable files (specifically Windows MZ executables) to common cloud storage services like Amazon S3, Google Cloud Storage, and Azure Blob Storage. It looks for HTTP PUT requests containing the 'MZ' magic bytes at the beginning of the file data, indicating a Windows executable, with a file size greater than 100KB, directed towards known cloud storage domains.
This rule detects potential watering hole attacks by identifying HTTP responses (flow to client) that contain a combination of HTML script tags, 'document.write', and 'eval(' within the file data. This pattern is indicative of malicious JavaScript injection designed to execute code on the client side.
This rule detects the download of common Remote Access Tools (RATs) such as AnyDesk, TeamViewer, UltraViewer, RustDesk, and Zoho Assist. It specifically looks for these keywords in the HTTP URI during client-to-server HTTP traffic. The rule is categorized as 'trojan-activity' and is intended to alert when a user might be downloading these tools, potentially after social engineering or phone contact, as suggested by the rule's message.
This rule detects HTTP responses from a C2 beacon that contain the keywords "kill" and "process" (case-insensitive) within the URI. This pattern suggests a command being sent from the command and control server to the client to terminate a process, potentially an antivirus process, as part of defense evasion tactics.
This rule detects HTTP traffic where the URI contains both 'disable' and 'defender' (case-insensitive). This pattern suggests an attempt to disable security tools, specifically Windows Defender, via an HTTP command. The flow is set to 'to_client,established', indicating the response from a server to a client within an established connection.
This rule detects attempts to remotely clear event logs over Server Message Block (SMB) by looking for the strings "Eventlog" and "ClearEventLog" in TCP traffic destined for port 445. This activity is often associated with adversaries attempting to remove forensic evidence of their presence on a system.
This rule detects network traffic indicative of WMI remote execution, specifically looking for the string "IWbemServices" on TCP port 135. This activity is often associated with Living Off The Land Binaries and Scripts (LOLBAS) techniques, where legitimate system tools like WMI are abused for malicious purposes such as remote code execution or lateral movement. The rule is designed to alert on potential trojan activity.
This rule detects the download of a COM scriptlet (.sct file) over HTTP that contains XML and the keyword 'scriptlet'. This behavior is characteristic of the 'Squiblydoo' technique, which abuses legitimate Windows binaries like Regsvr32 to execute arbitrary code, often for defense evasion.
This rule detects HTTP traffic where the User-Agent header contains the string 'CertUtil' (case-insensitive). This pattern is often associated with adversaries using the Windows built-in utility CertUtil to download malicious payloads, a common technique for defense evasion.
This rule detects HTTP downloads where the file is disguised as an image (e.g., .jpg, .png) but its content starts with 'MZ', indicating it's a Windows executable. This is a common technique used by adversaries to bypass security measures that might block executable downloads but allow image files.
Page 83 of 93