The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill possesses a vulnerability surface where it ingests and processes output from standard Linux networking diagnostic tools. Maliciously crafted resource names or service outputs could theoretically influence agent behavior. * Ingestion points: Command output from diagnostic tools including ip, ss, nmcli, resolvectl, and dig. * Boundary markers: The skill does not define specific delimiters or instructions to ignore embedded commands within tool output. * Capability inventory: The skill utilizes sudo for administrative commands and has the ability to modify system network configuration files. * Sanitization: No specific filtering or sanitization of external tool output is documented.
[SAFE]: The skill operates as a legitimate administrative resource. All privileged operations, such as modifying configuration files in /etc/netplan/ and executing network probes via sudo, are standard for the skill's intended purpose. The documentation emphasizes safety through the netplan try workflow, which provides an automatic rollback in case of connectivity loss.

linux-network-admin