Detections
Explore public detection logic contributed by the community across SIEM and rule languages.
2,588 detections
Filters
Last updated
All Time
Detection languages
1,194
459
310
204
161
Contributors
487
455
323
171
123
Categories
1,884
937
415
400
360
Platforms
1,349
560
343
329
257
Products / Services
601
314
306
286
271
MITRE Techniques
18,127
15,648
12,596
8,407
6,231
2,588
CVEs
9
5
5
5
2
IDS Classtypes
142
96
49
11
6
IDS Protocols
177
49
44
17
4
This rule detects devices making a high volume of successful network connections to GitHub's API endpoints (api.github.com or api.releases). A count of 5 or more successful connections within an hour from a single device is considered suspicious and could indicate data exfiltration, automated tooling, or unusual development activity.
This rule detects network connections to known cloud storage, file hosting, and webhook services (Discord, Pastebin, Dropbox, Telegram, AnonFiles) initiated by processes other than standard web browsers. Such behavior is often indicative of data exfiltration or C2 activity using legitimate web services to blend in with normal traffic.
Detects DNS queries to GitHub API domains originating from non-standard or potentially spoofed process names associated with various AI tools, which may indicate data exfiltration or unauthorized interaction with repositories.
Detects network connections initiated by Windows Nearby Sharing processes (quick_share.exe, nearby_sharing.exe, or NearbyShare.exe) to non-standard ports, potentially indicating unauthorized file transfers or exfiltration over unusual network channels.
Detects DNS queries to GitHub API domains originating from non-standard or potentially spoofed process names associated with various AI tools, which may indicate data exfiltration or unauthorized interaction with repositories.
Detects outbound web traffic directed towards common password management services (e.g., 1Password, LastPass, Bitwarden) originating from user agents associated with AI and LLM tools (e.g., OpenAI, Claude, Copilot). This combination of destinations and user agents is highly anomalous and may indicate potential credential exfiltration or interaction with automated agents acting on behalf of a user to access sensitive vault data.
This rule detects anomalous, high-frequency HTTP POST requests directed at sensitive API endpoints related to payment, billing, or administration. By calculating requests per minute and identifying unique source IP behavior, it aims to flag potential automated brute-force attacks, unauthorized API discovery, or data exfiltration attempts against critical system management interfaces.
Detects high-volume file writes or the use of common command-line copy utilities (xcopy, robocopy, copy) targeting removable drive paths (E-G:). This behavior is often indicative of data staging or exfiltration activities.
Detects the use of the FTP STOR command to transfer files with sensitive extensions (.docx, .pdf, .xlsx, .zip, .7z) to external networks. This behavior is indicative of potential data exfiltration via an unencrypted alternative protocol.
Detects repeated use of the FTP 'STOR' command to transmit data to an external host within a short timeframe. The STOR command is used in the File Transfer Protocol (FTP) to upload a file to the server. An unusual frequency of this command toward external destinations may indicate data exfiltration.
Detects data transfer to potentially attacker-controlled cloud accounts
Detects suspicious patterns of LLM API usage by monitoring outbound traffic to major LLM providers (OpenAI, Anthropic, Google, DeepSeek, Mistral, Cohere). The rule identifies potential API key theft ('LLMjacking') by flagging high request volumes from single keys, usage of the same key across multiple distinct source IP addresses, and direct outbound traffic bypassing established proxy/gateway configurations.
Detects HTTP traffic patterns consistent with data exfiltration to major cloud storage providers (AWS S3, Azure Blob Storage, Google Cloud Storage) using common command-line interface tools like rclone, aws-cli, and AzCopy.
Detects potential DNS-based command and control (C2) or exfiltration behavior by monitoring for DGA-like domain patterns, high-volume DNS requests, and the use of command-line tools (nslookup.exe, dnscmd.exe) to perform TXT record lookups often associated with data exfiltration.
This rule detects users uploading a high volume of files (50 or more within an hour) to SharePoint or OneDrive. This activity could indicate data exfiltration or unauthorized mass data transfer.
This rule detects when an account copies five or more files, each larger than 50MB, within a one-hour period. This activity could indicate data exfiltration or staging for exfiltration.
This rule detects when a single process account creates or copies 10 or more files to a removable media device (e.g., USB drive) within a one-hour window. This behavior can be indicative of data exfiltration attempts.
Detects indicators associated with the 'smart_window' campaign targeting Firefox, which attempts to exfiltrate browser history and page content via unauthorized registry modifications, malicious file artifacts, and specific network communication patterns to localhost or external URLs.
Detects anomalous, high-volume data transfers to Telegram API endpoints (api.telegram.org, specific IP ranges), which is indicative of data exfiltration by malware such as KuinaExtractor. The rule monitors for cumulative outbound traffic exceeding 500MB within a 1-hour session window per process.
This rule detects executions of 'winscp.exe' that are spawned by suspicious parent processes commonly associated with web servers or command interpreters (e.g., w3wp.exe, httpd.exe, powershell.exe). It also flags instances where WinSCP is invoked with command-line arguments indicative of automated scripts, custom configuration files, or remote transfer protocols (SFTP/SCP/FTP), which may indicate data staging or exfiltration activities.
Detects network connections attempting to access Kubernetes API endpoints related to secrets. This could indicate an adversary trying to enumerate or exfiltrate sensitive credentials from a Kubernetes cluster.
Page 49 of 124