Browse Rules

Potential EtherHiding C2 via Curl JSON-RPC Request

Detects when curl or nscurl is spawned by a shell or osascript to make a JSON-RPC request to a blockchain endpoint for command and control purposes. Adversaries may leverage blockchain smart contracts as a covert C2 channel to retrieve commands or payload staging information, as observed in ClickFix campaigns.

Launch Service Creation and Immediate Loading

An adversary can establish persistence by installing a new launch agent that executes at login by using launchd or launchctl to load a plist into the appropriate directories.

Potential Etherhiding C2 via Blockchain Connection

Detects when a scripting interpreter makes an outbound network connection to an Ethereum blockchain endpoint for command and control purposes. Adversaries may leverage Ethereum blockchain infrastructure as a covert C2 channel to receive commands and exfiltrate data, as observed in campaigns like SleepyDuck malware.

Suspicious Java Class File Created in PaperCut Server Library

Detects creation of Java .class files within the PaperCut NG/MF Application Server library directory. During active exploitation of CVE-2026-82078 (chained with CVE-2026-81578), attackers deliver hex-encoded malicious .class payloads into the PaperCut server/lib path (observed examples include Udydn.class and Moo97.class) so arbitrary bytecode executes inside the PaperCut JVM / Application Server process.

Potential Java Service Exploitation via Suspicious Child Process

Identifies a Java process that accepts an inbound network connection and then spawns a suspicious child process. This may indicate exploitation of a Java service that runs attacker-controlled code, such as one that deserializes untrusted objects.

Dumping Account Hashes via Built-In Commands

Identifies the execution of macOS built-in commands used to dump user account hashes. Adversaries may attempt to dump credentials to obtain account login information in the form of a hash. These hashes can be cracked or leveraged for lateral movement.

Finder Sync Plugin Registered and Enabled

Finder Sync plugins enable users to extend Finder’s functionality by modifying the user interface. Adversaries may abuse this feature by adding a rogue Finder Plugin to repeatedly execute malicious payloads for persistence.

Potential Privacy Control Bypass via TCCDB Modification

Identifies the use of sqlite3 to directly modify the Transparency, Consent, and Control (TCC) SQLite database. This may indicate an attempt to bypass macOS privacy controls, including access to sensitive resources like the system camera, microphone, address book, and calendar.

Remote SSH Login Enabled via systemsetup Command

Detects use of the systemsetup command to enable remote SSH Login.

Suspicious Web Browser Sensitive File Access

Identifies the access or file open of web browser sensitive files by an untrusted/unsigned process or osascript. Adversaries may acquire credentials from web browsers by reading files specific to the target browser.

Suspicious Process Execution by Zoom

Identifies suspicious process execution associated with the Zoom desktop client on macOS and Linux. The rule detects shells, script interpreters, downloaders, and network utilities spawned by Zoom on either platform. On Linux, it also detects Zoom replacing its own process image with an executable outside the Zoom installation directory. These behaviors may indicate successful exploitation of a Zoom client vulnerability, including CVE-2026-53413.

SSFileCopyReceiver Writing to Common Persistence Locations

Identifies the macOS Screen Sharing file copy helper SSFileCopyReceiver creating or modifying files in common persistence locations, including system-wide and per-user LaunchDaemons/LaunchAgents, shell profiles, SSH authorized_keys, cron tabs, and hidden paths under root's home directory. SSFileCopyReceiver performs file writes with root authority on behalf of a remote viewer.Pre-authentication exploitation of the Screen Sharing service (CVE-2026-65400) abuses this write primitive to establi

SSFileCopySender Executed as Root

Identifies execution of the macOS Screen Sharing file-copy helper SSFileCopySender with root UID/GID attributes (0 80). Under the native Apple authentication path this helper runs in the connecting user's context; execution as root is anomalous and consistent with pre-authentication exploitation of the Screen Sharing service (CVE-2026-65400), where a flawed SRP validation path lets an unauthenticated attacker reach privileged file operations. Note that the 0/80 UID/GID pair reflects only initial

Elastic Agent Service Terminated

Identifies the Elastic endpoint agent has stopped and is no longer running on the host. Adversaries may attempt to disable security monitoring tools in an attempt to evade detection or prevention capabilities during an intrusion. This may also indicate an issue with the agent itself and should be addressed to ensure defensive measures are back in a stable state.

GenAI or MCP Server Child Process Execution

Detects child process execution from GenAI tools or MCP (Model Context Protocol) servers. Adversaries exploit AI agents to execute system commands, exfiltrate data, or establish persistence. MCP servers provide LLMs direct access to execute shell commands, read files, and interact with external services. This building block provides visibility into AI-initiated process execution for correlation with other suspicious activity.

LLM-Based Curl Activity Triage

Detects non-allowlisted curl activity on Linux, macOS, and Windows hosts and uses an LLM to assess whether the activity is malicious, benign, or requires investigation. The rule parses and normalizes the destination, redacts sensitive command-line values, and aggregates activity by host and destination before invoking the ES|QL COMPLETION command. Only true positive or suspicious verdicts with confidence above 0.7 generate alerts.

LLM-Based Wget Activity Triage

Detects non-allowlisted wget activity on Linux, macOS, and Windows hosts and uses an LLM to assess whether the activity is malicious, benign, or requires investigation. The rule parses and normalizes the destination, redacts sensitive command-line values, and aggregates activity by host and destination before invoking the ES|QL COMPLETION command. Only true positive or suspicious verdicts with confidence above 0.7 generate alerts.

Claude Cowork VM Boot Image Tamper

Detects unexpected modification of Claude Desktop Cowork VM boot images (kernel, initrd, root filesystem). Adversaries with user-context access can rewrite these stored images so later Cowork sessions boot attacker-controlled code inside a virtual instance that host EDR cannot inspect by default.

Potential Tunneling via Tailscaled

Identifies the use of Tailscaled to potentially tunnel network traffic. This can be used by attackers to enable routing of network packets that would otherwise not reach their intended destination, or to bypass network restrictions and/or hide traffic from network monitoring.

Suspicious Curl from macOS Application

Detects the use of curl by a macOS application binary to connect to a raw IP URI and download a second stage payload. Threat actors often utilize a benign looking or legitimate application as a first stage dropper. Curl is commonly used as it doesn't enforce Gatekeeper checks.

Emond Rules Creation or Modification

Identifies the creation or modification of the Event Monitor Daemon (emond) rules. Adversaries may abuse this service by writing a rule to execute commands when a defined event occurs, such as system start up or user authentication.

External IP Address Discovery via Curl

Detects applications making a curl request to a known public IP address lookup web service. Malware commonly performs this action during reconnaissance to assess potential targets and identify the victim's external IP address.

Potential Data Exfiltration Through Curl

Detects the use of curl to upload files to an internet server. Threat actors often will collect and exfiltrate data on a system to their C2 server for review. Many threat actors have been observed using curl to upload the collected data. Use of curl in this way, while not inherently malicious, should be considered highly abnormal and suspicious activity.

New USB Storage Device Mounted

Identifies newly seen removable devices by device.serial_number and host.id using the Elastic Defend device mount events. While this activity is not inherently malicious, analysts can use those events to aid monitoring for data exfiltration over those devices.

GenAI Process Connection to Unusual Domain

Detects GenAI tools connecting to unusual domains on macOS. Adversaries may compromise GenAI tools through prompt injection, malicious MCP servers, or poisoned plugins to establish C2 channels or exfiltrate sensitive data to attacker-controlled infrastructure. AI agents with network access can be manipulated to beacon to external servers, download malicious payloads, or transmit harvested credentials and documents.