Detections
Explore public detection logic contributed by the community across SIEM and rule languages.
20,501 detections
Filters
Last updated
All Time
Detection languages
8,116
7,174
1,351
1,169
1,063
Contributors
4,878
2,724
2,695
2,646
2,356
Categories
20,501
11,881
5,887
5,179
4,868
Platforms
18,096
3,056
1,506
124
104
Products / Services
5,727
4,303
3,717
1,175
958
MITRE Techniques
9,493
7,577
3,461
2,848
2,703
CVEs
36
28
25
21
20
IDS Classtypes
1
This rule detects the execution of known Remote Monitoring and Management (RMM) or remote access tools on devices where they have not been observed within the previous 90-day baseline (excluding the most recent 7 days). This pattern is often indicative of an adversary introducing unauthorized remote access capabilities for persistence or lateral movement.
Detects curl liveness/health-check requests against a known tunnel domain issued repeatedly from a loop construct or a LaunchAgent/watchdog parent, indicating an automated watchdog keeping a reverse tunnel alive.
Detects pgrep searching for ngrok, cloudflared, or wireguard-go process names from within a Claude Code shell-snapshot session, indicating discovery of tunnel/VPN tooling as part of an agent-driven attack chain.
Detects the use of native Windows utilities such as vssadmin, wbadmin, wmic, diskshadow, and bcdedit to delete volume shadow copies, backup catalogs, or modify boot configuration to prevent system recovery. The rule explicitly excludes designated backup servers to reduce noise from legitimate management activities.
Detects a curl request to a tunnel-exposed /api/summary endpoint piped through jq extracting a 'spend' field, indicating automated exfiltration of financial data via a coding-agent-established tunnel.
Detects PowerShell command lines that fetch a remote module using methods like DownloadString or Invoke-WebRequest, subsequently executing the downloaded code in-memory via Invoke-Expression (iex), and finally invoking a main() function. This pattern matches the execution chain used by the CountLoader malware (taskType 5).
Detects instances where a Java process (typically a web application server like Tomcat) initiates suspicious command-line tools such as cmd.exe, powershell.exe, or various Linux shell binaries, using command-line arguments indicative of a reverse shell or code execution (e.g., -e /bin/sh, -i, /dev/tcp, IEX, DownloadString). This pattern is commonly associated with exploit attempts targeting vulnerabilities like Java deserialization.
This rule detects the execution of QuickAssist.exe or RemSupp.exe on a host where Microsoft Teams has also been executed within the last day. This pattern may indicate the use of built-in remote support tools in conjunction with communication platforms, potentially for unauthorized remote access or suspicious remote assistance activities.
This rule detects hosts exhibiting behavior consistent with an active ransomware or destructive attack in progress, by correlating five distinct attack-chain stages within a rolling 24-hour window: environment/AD enumeration, network share discovery, lateral movement via RDP, credential dumping, and backup/recovery destruction.
This rule detects potential privilege escalation through the abuse of a debug backdoor in the Wacom WTabletServiceISD service. It identifies when the 'PowerT' registry value is set for the service, followed shortly by the service spawning a command shell (cmd.exe or powershell.exe) under SYSTEM privileges. It specifically excludes scenarios where AutoAdminLogon is enabled, which might otherwise trigger false positives.
This rule detects potential credential theft activity where a process spawning a command-line interpreter (bash, sh, zsh) using 'curl' simultaneously accesses sensitive browser configuration or credential directories on MacOS, such as Chrome, Firefox, or Safari application support paths. This pattern is indicative of an adversary attempting to exfiltrate browser-stored secrets.
Detects process activity referencing GitHub Actions, GitLab CI, or Azure DevOps pipeline environment variable names (GITHUB_*, CI_*, GITHUB_ACTIONS, GITLAB_CI, Azure DevOps pipeline variables) consistent with a malicious extension harvesting CI/CD runner environment variables.
This rule detects the potential replacement of legitimate cryptocurrency wallet applications (Ledger, Trezor, Exodus) by monitoring for the download of zip archives from suspicious paths containing 'zxc' using curl, followed by the creation or modification of files within known wallet application directories. This activity is indicative of a trojanized application delivery mechanism, often associated with info-stealer campaigns.
This rule detects PowerShell execution patterns indicative of fileless information stealer activity. It searches for specific command-line flags often associated with obfuscated or hidden script execution (e.g., -EncodedCommand, -nop -w hidden, IEX) combined with attempts to access or target sensitive files and tokens (e.g., Cookies, Login Data, msal.cache, AAD.BrokerPlugin) common in Microsoft 365 and browser-based credential theft.
This rule identifies non-Microsoft processes (based on file name, file company name, and process signer) that initiate network connections to known Microsoft authentication endpoints. This behavior is often indicative of an adversary-controlled process attempting to leverage cloud services or perform authentication-related activities outside of expected system or productivity software.
Detects the execution of processes related to 'SKILL.md', 'npx', 'mcp-server', or 'invoke_skill' followed closely by read or copy actions on sensitive file locations (SSH keys, AWS credentials, environment files, or AppData). The rule further correlates this activity with subsequent outbound network connections from the same device.
Detects Apple-signed python/python3 or osascript referencing the same /tmp/*.js file path twice in one command line (drop-then-execute), indicating LOLBin abuse to stage and immediately re-run a JavaScript payload from a world-writable directory.
Detects a decrypted Keychain dump (security dump-keychain -d) spawned from a Cursor/coding-agent process with output filtered for Linear, MCP, or OAuth-related strings, indicating targeted credential theft of MCP/Linear integration tokens.
Detects the macOS security CLI dumping the Claude Code OAuth credential item from Keychain (find-generic-password -w targeting 'Claude Code-credentials') by a process other than Claude Code itself, indicating credential theft targeting the coding agent's stored token.
Detects process self-deletion (executable, scheduled task, or install directory) correlated within 60 minutes to a matching prior creation event, indicating post-execution cleanup behavior consistent with loader/RAT install-then-delete patterns such as those used by DOUBLECUP-delivered payloads.
Detects signs of Cloudflared tunnel persistence on Windows endpoints, specifically targeting service registration (via sc.exe or registry modifications) and the presence of the cloudflared.exe binary. This is often used by attackers to maintain outbound-only access to compromised machines following exploitation of vulnerabilities like CVE-2026-18577.
Page 298 of 977