IoTNet
Warn
Audited by Gen Agent Trust Hub on Feb 26, 2026
Risk Level: MEDIUMCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes the
iotnetbinary to perform analysis on network traffic files and live interfaces. - [COMMAND_EXECUTION]: The instructions explicitly direct the agent to use
sudofor live network captures, granting the underlying tool elevated root-level privileges on the system. - [PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection by ingesting and processing untrusted data from network packets.
- Ingestion points: External PCAP files and live network traffic data entering the agent's context through tool output.
- Boundary markers: There are no instructions or delimiters defined to separate the potentially malicious content of network packets from the agent's core instructions.
- Capability inventory: The agent can execute shell commands with elevated privileges and access the local file system for PCAP files.
- Sanitization: No sanitization or validation of the packet content or capture filters is implemented before processing.
Audit Metadata