The Agent Skills Directory

[Privilege Escalation] (HIGH): The skill explicitly instructs the agent to execute commands using sudo for live network captures (e.g., sudo iotnet -i eth0 -d 30). This grants the tool root-level access to the host system.
[Unverifiable Dependencies] (MEDIUM): The skill relies on an external binary located in the iothackbot directory. The source, integrity, and safety of this 'iothackbot' suite cannot be verified, which is a major security risk when combined with elevated privileges.
[Indirect Prompt Injection] (HIGH): This skill exhibits a high-risk attack surface for indirect prompt injection.
Ingestion points: Processes untrusted network traffic and packet capture (PCAP) files.
Boundary markers: None present; the skill treats all packet data as legitimate input for its analysis engine.
Capability inventory: Executes shell commands with sudo privileges via the iotnet binary.
Sanitization: No evidence of sanitization for packet contents or BPF filters provided by users.
[Command Execution] (HIGH): The skill executes arbitrary commands on the host system, including parameters like --config which could potentially be used to point to malicious configuration files or trigger path traversal vulnerabilities.

iotnet