linux-networking

Warn

Audited by Gen Agent Trust Hub on Mar 7, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONCREDENTIALS_UNSAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill requires administrative privileges for its core operations, specifically instructing the use of sudo for tasks like netplan apply, ufw enable, and wg-quick.
  • [CREDENTIALS_UNSAFE]: The skill manages sensitive authentication data, including Wireguard private keys and Tailscale API keys. While it recommends using environment variables, the agent still processes and handles these secrets directly.
  • [EXTERNAL_DOWNLOADS]: The skill documentation includes references to the official Tailscale API for managing network devices and peer connectivity.
  • [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface by processing external configuration data.
  • Ingestion points: Reads configuration files such as /etc/netplan/.yaml and /etc/wireguard/.conf.
  • Boundary markers: None specified in the instructions for parsing or validating these files.
  • Capability inventory: High-privilege command execution via sudo, including network configuration changes and service management.
  • Sanitization: No explicit validation or sanitization of the configuration content before execution is mentioned beyond the use of yamllint for syntax checking.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Mar 7, 2026, 05:45 PM