Detections
Explore public detection logic contributed by the community across SIEM and rule languages.
3,624 detections
Filters
Last updated
All Time
Detection languages
1,749
374
349
273
255
Contributors
908
661
593
359
164
Categories
1,486
1,150
779
525
515
Platforms
39,894
7,228
6,523
3,864
3,624
Products / Services
802
512
432
340
297
MITRE Techniques
1,177
572
523
506
437
CVEs
17
16
9
8
7
IDS Classtypes
41
15
10
6
5
IDS Protocols
47
24
3
2
1
Detects Node.js executing the malicious setup.cjs script with the --no-warnings flag. This execution sequence is associated with the easy-day-js typosquat postinstall hook used in the Mastra npm supply chain compromise.
This rule detects network activity associated with the StealC information stealer, specifically focusing on HTTP POST requests made to known command-and-control (C2) domains. It identifies characteristic URL patterns, such as randomly generated 16-character PHP files or specific index.php structures, which are used for bot registration and communication. The detection includes patterns indicative of RC4-encrypted payloads and behavior mimicking Amadey bot status checks.
Detects execution of the macOS 'security' utility to harvest generic or internet passwords from the Keychain, a technique commonly associated with the AMOS (Atomic macOS Stealer) infostealer for credential theft.
Detects the AMOS (Atomic macOS Stealer) infection chain where a shell (e.g., Terminal, bash) executes curl to download a DMG file to a temporary directory, followed immediately by the use of hdiutil with the -nobrowse flag to mount the disk image silently, effectively bypassing macOS Gatekeeper UI prompts to the user.
This rule detects a high frequency of successful network connections from a single device to multiple unique destination IP addresses within a one-hour window. The targeted ports (445, 139, 3389, 22, 21) are commonly used for remote administrative access, file transfers, and remote shell services, which are frequently leveraged by adversaries for lateral movement. The threshold of 3 unique targets in an hour may indicate automated scanning or credential brute-forcing activity across the network.
Detects a user mounting a deceptive .dmg file that invokes a hidden .sh or .command script upon opening. This rule specifically looks for the 'hdiutil' process attaching a disk image and then correlates it with a process executing a hidden shell script (.sh or .command) from a mounted volume.
Detects non-browser applications (Python, PowerShell, cURL, wget, node) communicating directly with OpenAI or ChatGPT API endpoints, indicative of automated AI tradecraft by threat actors. This rule specifically looks for network connections to api.openai.com initiated by these common scripting or command-line tools.
Detects network activity associated with the DcRAT trojan as identified in Operation DragonReturn. This includes outbound network connections to hardcoded C2 port 2671, connections to known malicious IP addresses, and DNS resolutions of identified attacker-controlled C2 domains.
Detects anomalous network traffic patterns communicating with the Telegram Bot API (api.telegram.org) from non-official Telegram processes. The rule identifies potential command and control (C2) activity by analyzing request frequency, specific API method usage (e.g., getUpdates, sendMessage), and beaconing-like behavior.
Detects Chrome launch with specific crash recovery evasion flags indicative of FlutterShell.
Detects activity associated with a specific phishing campaign distributing 'Distribution Settlement Agreement' PDF lures. The rule monitors for file creation or execution of files matching this naming pattern, as well as network connection attempts to domains identified as related to the threat actors: 'gasparinutrition.com' and 'gappssmtp.com'.
Detects network requests to Google image search containing an 'imgurl' parameter, which can be used by an attacker to exfiltrate data by forcing an external service to crawl a controlled URL containing sensitive information.
Detects the execution of sensitive command-line tools or environment variables (e.g., AWS CLI, MCP tools, environment secrets) triggered by AI-based development assistants such as Claude, Cursor, or Windsurf. This behavior may indicate an attacker abusing the permissions or capabilities of an AI coding assistant to perform unauthorized cloud discovery or credential access.
This rule detects potentially malicious child processes spawned by Amazon Q or AWS Language Server components (e.g., amazon-q, aws-lsp-server). The rule specifically flags behavior associated with the loading of MCP server configurations, potential unauthorized access to AWS credentials, or network data exfiltration attempts through commonly abused binaries like shell interpreters, python, or curl/wget when these are not signed by Amazon.
Detects the presence or activity of a known malicious Chromium-based browser extension (Perplexity AI imposter, ID: flkebkiofojicogddingbdmcmkpbplcd) through monitoring file system changes in browser directories, registry modifications, or network communication with identified command-and-control infrastructure.
Detects instances where common development runtime processes (e.g., node, python, shells) spawn child processes whose command lines contain references to sensitive credential environment variables such as AWS access keys, GitHub tokens, or SSH authentication sockets. This behavior often indicates that a developer or an automated tool is insecurely passing credentials via command-line arguments to subprocesses.
Detects the execution of AWS CLI commands by common scripting interpreters (Node.js or Python). This behavior is often indicative of automated cloud environment reconnaissance or credential testing using compromised access keys or instances.
Detects macOS Terminal (bash/zsh) retrieving loader.sh or payload.applescript directly from the ClickFix attacker infrastructure.
This rule detects network connections to Discord's API webhooks endpoint, specifically looking for POST requests. This activity can indicate data exfiltration, command and control communications, or other malicious use of Discord webhooks by adversaries.
Detects Node.js executing the malicious setup.cjs script with the --no-warnings flag. This execution sequence is associated with the easy-day-js typosquat postinstall hook used in the Mastra npm supply chain compromise.
Detects processes attempting to read or open sensitive SSH files (keys and configuration) that are not recognized as legitimate SSH-related utilities (e.g., ssh, sshd, rsync, git). This behavior is often indicative of credential harvesting or unauthorized reconnaissance of authentication artifacts.
Page 111 of 173