Detections
Explore public detection logic contributed by the community across SIEM and rule languages.
2,431 detections
Filters
Last updated
All Time
Detection languages
2,428
3
Contributors
747
554
405
315
183
Categories
1,914
1,291
689
671
449
Platforms
2,202
183
149
62
49
Products / Services
2,366
1,378
38
37
25
MITRE Techniques
537
454
342
308
301
CVEs
17
10
9
9
9
IDS Classtypes
1,098
452
204
154
102
IDS Protocols
2,431
868
408
324
97
This rule detects attempts to use cloud services (specifically AWS SSM or Azure Management) for internal lateral movement. It triggers when HTTP requests are made to 'ssm.amazonaws.com' or 'management.azure.com' with the URI containing 'SendCommand'. A threshold of 5 such events from the same source within 30 seconds is set to reduce false positives.
This rule detects potential session hijacking attempts by monitoring HTTP traffic for unusually long cookie values. Specifically, it looks for HTTP cookies containing base64-like strings (alphanumeric characters, plus, slash, and equals signs) that are 100 characters or longer. A threshold of 5 such occurrences within 3 seconds from the same destination IP address triggers the alert, indicating a possible attempt to use a stolen browser session cookie.
This rule detects potential browser session token injection via a Man-in-the-Middle (MitM) attack. It looks for HTTP responses (flow:to_client) that contain the strings "document.cookie", "=", and "XMLHttpRequest" (case-insensitive) within the file data. The combination of these strings in a client-bound HTTP response suggests an attempt to manipulate browser cookies, possibly to inject session tokens, often seen in MitM scenarios where an attacker intercepts and modifies traffic to compromise user sessions.
This rule detects the exfiltration of a custom XOR-compressed archive over HTTP. It specifically looks for HTTP POST requests originating from the home network, containing file data, with a payload size greater than 10000 bytes, and a PCRE match for a sequence of at least 200 non-ASCII or control characters, which is indicative of XOR-compressed binary data.
This rule detects when a user downloads a large amount of Slack message history via the Slack API. It specifically looks for HTTP requests to 'slack.com' with the URI '/conversations.history' and triggers an alert if 20 such requests are observed from the same source IP within a 60-second window. This behavior could indicate an attempt to exfiltrate sensitive communication data.
This rule detects the exfiltration of browser bookmarks over HTTP POST requests. It specifically looks for HTTP POST requests where the client body contains the string "Bookmarks" (case-insensitive) and the data size is greater than 500 bytes. This pattern is indicative of an attacker or malware attempting to steal a user's browser bookmarks.
This rule detects HTTP requests to the Kubernetes API endpoint '/api/v1/pods' on port 8080. This activity is indicative of an attempt to discover Kubernetes pods and services, which could be part of reconnaissance by an attacker or an authorized user.
This rule detects potential sandbox evasion techniques by identifying HTTP traffic (specifically client-to-server responses) that contains both 'GetCursorPos' and 'Sleep' within the file data. This combination is often used by malware to check for user interaction (mouse movement) and introduce delays, which can indicate an attempt to determine if it's running in a sandboxed environment.
This rule detects network traffic (specifically HTTP responses) that contains patterns indicative of malware attempting to prevent double execution using mutexes. It looks for the strings "CreateMutex" and "ALREADY_EXISTS" (case-insensitive) within the file data of the HTTP response. This behavior is common in malware to ensure only one instance of itself is running on a system, often as a defense evasion technique.
This rule detects the download of executable files (EXE, DLL, BIN) over HTTP that appear to be stripped of their standard headers. It specifically looks for the 'MZ' magic bytes at the beginning of the file data, indicating a Windows executable, but then implies a lack of full header structure by the 'Stripped Executable Payload Without Headers' message. The rule also requires the file size to be greater than 10000 bytes.
This rule detects HTTP responses containing SVG (Scalable Vector Graphics) content that also embeds base64 encoded data starting with 'TVQAA'. This specific sequence ('TVQAA') is characteristic of a base64 encoded Windows executable (MZ header). The rule aims to identify a technique known as SVG smuggling, where malicious executables are hidden within SVG files and delivered to clients, often as a defense evasion tactic.
This rule detects HTTP responses that contain a URI with a double extension, where the first extension is a common document or image file type (doc, pdf, jpg, png, txt) and the second extension is an executable or script file type (exe, bat, vbs, ps1, cmd). This pattern is often used in file masquerading to trick users into executing malicious files.
This rule detects attempts to submit a Dockerfile containing 'FROM' and 'RUN' instructions to the Docker API via an HTTP POST request to the '/build' endpoint. This activity could indicate an attacker attempting to build a malicious Docker image on a compromised Docker host.
This rule detects the delivery of a CMSTP INF file over HTTP that contains specific content indicative of a User Account Control (UAC) bypass attempt. It looks for HTTP responses containing a URI with ".inf", and within the file data, the strings "[DefaultInstall]" and "RegisterOCXs". This combination suggests an attempt to leverage CMSTP (Connection Manager Profile Installer) to execute arbitrary code, often used by adversaries for defense evasion and privilege escalation.
This rule detects the download of a Microsoft Management Console (MMC) snap-in file (.msc) over HTTP that also contains the string "MMCVersion". This combination is often used by attackers to execute malicious code or establish persistence, leveraging the trusted nature of MMC snap-ins. The rule specifically looks for the file extension in the URI and the 'MMCVersion' string within the file's data, indicating a potential malicious MMC snap-in being delivered via HTTP.
This rule detects HTTP requests that contain both 'otp' and 'verify' in the URI, which is indicative of a phishing page designed to capture One-Time Passwords (OTP) or Multi-Factor Authentication (MFA) codes. The rule is specifically looking for patterns associated with reverse proxy phishing techniques.
This rule detects potential exfiltration of password manager database files over HTTP. It specifically looks for HTTP POST requests with a URI containing common password manager names (keepass, lastpass, 1password, bitwarden) and a data size greater than 1000 bytes. This combination suggests a large file, potentially a database, is being uploaded.
This rule detects attempts to harvest Slack messages for credential tokens by monitoring HTTP requests to slack.com with the URI '/conversations.history'. A threshold of 20 such requests within 60 seconds from the same source IP triggers the alert, indicating potential automated data exfiltration or reconnaissance.
This rule detects the submission of a Golden SAML attack token to a service. It specifically looks for HTTP POST requests where the URI contains 'saml', the client body contains 'Assertion', and the body size is greater than 2000 bytes. This combination of indicators suggests an attempt to authenticate using a forged SAML assertion.
This rule detects the use of a forged SAML token for identity federation bypass. It specifically looks for HTTP POST requests (flow:to_server,established; http.client_body;) containing the string "SAMLResponse" (content:"SAMLResponse"; nocase;) and a long base64-encoded string (pcre:"/[A-Za-z0-9+\/]{500,}={0,2}/";) within the client body. The presence of a large base64 string in conjunction with "SAMLResponse" suggests a potentially malicious SAML assertion being submitted.
This rule detects the delivery of a Linux ptrace injection tool via HTTP. It specifically looks for the string "ptrace(PTRACE_ATTACH" within HTTP response data, indicating that a file containing this specific ptrace function call is being sent to a client. This technique is often used for process injection and privilege escalation on Linux systems.
Page 74 of 116