The Agent Skills Directory

[PROMPT_INJECTION] (SAFE): No instructions were found that attempt to bypass AI safety protocols or override system instructions. The instruction 'Do This First' is used as a task prioritization marker for reading the provided reference file.
[DATA_EXFILTRATION] (SAFE): No sensitive file paths, credential markers, or network transmission patterns were detected. The skill actively promotes security best practices by advising users never to pass secrets via CLI flags.
[REMOTE_CODE_EXECUTION] (SAFE): The skill is purely documentation-based and does not include any logic for downloading or executing remote code. It explicitly states to avoid implementation and stick to designing parameters.
[COMMAND_EXECUTION] (SAFE): No shell command execution, subprocess spawning, or binary calls are present in the skill files.
[EXTERNAL_DOWNLOADS] (SAFE): The referenced URLs (clig.dev and GitHub) are static links for user documentation. No automated tools for downloading or installing external packages were found.

create-cli