The Agent Skills Directory

[SAFE]: The skill is purely instructional, providing templates and guidelines for organizing markdown files and metadata. It does not perform any network operations or access sensitive local data.
[PROMPT_INJECTION]: No malicious override instructions or jailbreak attempts were detected. The instructions follow standard formatting for agent guidance.
[COMMAND_EXECUTION]: While the skill provides examples of how an agent might use scripts (e.g., 'python scripts/validate.py'), it does not execute any commands itself and does not include suspicious shell scripts.
[DATA_EXFILTRATION]: There are no indicators of data being sent to external servers or access to sensitive files like credentials or SSH keys.

create-skill