Detections
Analytics
Command and Scripting Interpreter
Description
Adversaries may abuse command and script interpreters to execute commands, scripts, or binaries. AI Agents that wield tools effectively act as scriptable execution engines: prompt injection or compromised instructions can drive an agent to invoke its registered tools to perform attacker-chosen actions on behalf of the agent's identity.
Products, Sensors, and Dependencies
| Product | Dependencies | Data source | Access required | Protocol | Data Collected | Notes |
|---|---|---|---|---|---|---|
| Tenable AI Aware | AI | Read-only | HTTPS | AI Agents and Tools |
Attack Path Technique Details
Framework: MITRE ATT&CK
Family: Execution
Technique: Command and Scripting Interpreter
Products Required: Tenable AI Aware