Detections
Explore public detection logic contributed by the community across SIEM and rule languages.
7,128 detections
Filters
Last updated
All Time
Detection languages
3,206
1,086
504
478
469
Contributors
1,824
1,607
872
505
300
Categories
2,966
2,465
1,059
1,020
903
Platforms
39,752
7,128
6,517
3,846
3,595
Products / Services
1,324
1,059
969
657
487
MITRE Techniques
2,223
1,317
1,309
872
870
CVEs
66
65
33
32
28
IDS Classtypes
81
80
56
27
10
IDS Protocols
149
98
6
5
4
This rule detects processes making a high volume of successful network connections (3 or more within an hour) to domains containing 'codeartifact' or 'artifactory'. This behavior could indicate an attempt to download tools, libraries, or other artifacts, potentially as part of an ingress tool transfer or supply chain attack.
This rule detects when a single process rapidly creates or modifies three or more YAML files (files ending with '.yaml' or '.yml') within an hour. It specifically looks for these activities in paths containing 'config' or 'pipeline', which could indicate configuration changes, deployment activities, or potential malicious manipulation of system or application settings.
This rule detects the creation of 'requirements.txt' or 'pom.xml' files on a system. These files are commonly used in software development projects (Python and Java/Maven, respectively) to manage dependencies and project configurations. Monitoring their creation can help identify legitimate development activities, but also potential malicious actions such as supply chain compromise where an adversary might introduce or modify project dependencies.
This rule detects when a single process account initiates a high volume (10 or more) of file move operations to folders associated with cloud storage services (e.g., 'cloud', 'dropbox', 'drive') within a one-hour window. This behavior could indicate data exfiltration or staging for exfiltration.
Bun Runtime Remote Download
Cortex XDR
Detects network connections downloading the Bun runtime from the oven-sh GitHub repository, a behavior exhibited by the Hades loader. This rule specifically looks for HTTP network events where the URL contains 'github.com' and 'oven-sh/bun/releases/download', indicating an attempt to download the Bun runtime.
Payload Ransomware File Artifacts
Cortex XDR
Detects the creation of files with the .payload extension or the RECOVER_payload.txt ransom note, indicating active encryption by Payload ransomware.
Detects the use of the GNU Debugger (gdb) to dump process memory (gcore) to disk, a technique abused by attackers to extract credentials residing in Runner.Worker. The rule specifically looks for process creation events where 'gdb' is in the command line along with process ID flags ('-p' or '--pid') and memory dumping commands ('dump memory' or 'gcore').
Detects enumeration of git credential helpers via 'git config --global credential.helper'. This command is used by adversaries, such as Miasma, to determine how credentials are stored (e.g., wincred, osxkeychain, store) before attempting extraction.
Detects instances where 'wget' is used to download content and pipe it directly into a shell interpreter (bash, sh, or zsh). This technique is often used for fileless execution, allowing adversaries to download and execute malicious scripts without writing them to disk.
Detects tampering with 'binding.gyp' files by processes other than 'npm', 'node', or 'yarn'. Modifying these build scripts can allow attackers to achieve arbitrary code execution during the 'npm install' phase, indicating a potential supply chain compromise.
Recent reporting indicates the campaign targeted GitHub repositories by inserting malicious GitHub Actions / CI workflows designed to steal secrets and credentials.
5,561 repositories were affected in roughly six hours
Attackers pushed 5,718 malicious commits.
The objective was credential theft from CI/CD environments, including:
cloud credentials
SSH keys
access tokens
CI/CD secrets
The campaign reportedly used:
fake automation-style commit identities such as build-bot, ci-bot, or similar
malicious .github/workflows changes
workflow triggers that either:
executed automatically on push / pull request, or
remained dormant until manually triggered
Web reporting further states the payload collected environment variables, cloud metadata credentials, and repository runner secrets, then exfiltrated them to 216.126.225.129:8443.
5,561 repositories were affected in roughly six hours
Attackers pushed 5,718 malicious commits.
The objective was credential theft from CI/CD environments, including:
cloud credentials
SSH keys
access tokens
CI/CD secrets
The campaign reportedly used:
fake automation-style commit identities such as build-bot, ci-bot, or similar
malicious .github/workflows changes
workflow triggers that either:
executed automatically on push / pull request, or
remained dormant until manually triggered
Web reporting further states the payload collected environment variables, cloud metadata credentials, and repository runner secrets, then exfiltrated them to 216.126.225.129:8443.
Detects the creation or modification of suspicious web-related script files (PHP, ASPX, JSP, SHTML) in common web server directories, excluding activities from legitimate web server processes. This rule aims to identify potential webshell deployment or other post-exploitation activities, specifically linked to APT28's targeting of Roundcube webmail servers.
Detects processes making network requests to the '/api/css.js' endpoint, which is indicative of the ErrTraffic malware.
This rule detects the spawning of a root shell from specific SUID binaries (su, sudo, doas, passwd, chsh) when the initiating process was run by an unprivileged user. This behavior is indicative of a privilege escalation exploit, potentially related to CVE-2026-45257 (BUMSRAKETE).
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.
Detects the creation of specific AI assistant configuration files (e.g., Cursor, Claude, Gemini, VS Code) by alternative runtimes. The Miasma malware targets modern AI coding assistants to inject hidden backdoors that trigger automatically when projects are opened.
This rule detects outbound network connections to URLs associated with common file sharing services (e.g., Dropbox, Mega, Gofile) or temporary email providers (e.g., Temp-Mail). Such connections can indicate data exfiltration attempts, use of unsanctioned services, or communication with command and control infrastructure.
This rule watches for Node or similar runtimes scanning cloud secret stores after installing a dependency. Stolen secrets can lead to full cloud account takeover. Validate the package against your software bill of materials. Ask the person who used the computer whether they installed new software or clicked a link, and involve your security team if the answer is unclear.
This rule looks for a package writing a short-lived JavaScript file with a p-prefix name in temp during install. That pattern matches known stealers and wipers. Quarantine the package version. Ask the person who used the computer whether they installed new software or clicked a link, and involve your security team if the answer is unclear.
This rule catches a preinstall script running an obfuscated index.js before the main install completes. Preinstall runs automatically without user consent. Never ignore preinstall additions in dependency updates. Ask the person who used the computer whether they installed new software or clicked a link, and involve your security team if the answer is unclear.
This rule detects suspicious downloads of Python packages from pypi.org or pypi.python.org by pip or python executables. It specifically looks for a high volume (5 or more) of downloads of common package names (numpy, requests, pandas, util, common) within a one-hour window from a single device. This could indicate an attempt to compromise the software supply chain by introducing malicious versions of popular packages.
Page 260 of 340