The Agent Skills Directory

[COMMAND_EXECUTION]: The skill defines patterns for executing commands through a caching wrapper (x ccmd). While this involves running shell commands like curl, it is the core intended functionality for providing efficiency in agent workflows.
[PROMPT_INJECTION]: The skill involves an indirect prompt injection surface. Ingestion points: Data enters the context via the stdout of commands (e.g., curl) executed through x ccmd (SKILL.md). Boundary markers: None are specified in the instructions to separate cached command output from agent instructions. Capability inventory: The skill uses x ccmd to execute arbitrary subprocesses (SKILL.md). Sanitization: No validation or sanitization of command output is described before the agent processes it.

ccmd