The Agent Skills Directory

[COMMAND_EXECUTION]: The skill provides the agent with access to a wide range of powerful system commands through the x-cmd suite. This includes process monitoring (x ps), network host discovery (x ip map), TCP port scanning (x ip tcp-portscan), and interactive storage analysis utilities (x df).
[PRIVILEGE_ESCALATION]: Several tools within the skill manage sensitive system security and administrative functions. These include TouchID authentication integration (x mac tidsudo), firewall management (x mac fw), SSH server configuration (x mac sshd), and automated root privilege handling for hardware health monitoring (x smart).
[EXTERNAL_DOWNLOADS]: The x ip geolite utility performs network requests to external IP geolocation services to retrieve metadata for specific IP addresses.
[INDIRECT_PROMPT_INJECTION]: The skill creates an attack surface by encouraging the agent to pipe raw, potentially untrusted system data (such as process names or disk metadata) into LLMs for analysis (e.g., using @gemini).
Ingestion points: System output from x ps, x smart, and other administration modules described in SKILL.md.
Boundary markers: Absent; the skill lacks instructions for the agent to use delimiters or ignore instructions that might be embedded in the system data being processed.
Capability inventory: Extensive system access including process management, file system oversight, and network utility control across all modules.
Sanitization: Absent; no methods for escaping or validating system output are mentioned before interpolation into AI prompts.

x-system