The Agent Skills Directory

SAFE (SAFE): No malicious instructions, obfuscation, or exfiltration patterns were detected across any of the files.- COMMAND_EXECUTION (INFO): The provided scripts are safe local utilities. package_skill.py automates the creation of zip-based skill packages, and quick_validate.py performs metadata validation using yaml.safe_load() to prevent unsafe YAML deserialization.- PROMPT_INJECTION (SAFE): Documentation in the references/ directory provides structural templates for the agent but does not contain bypass instructions or adversarial prompts.

skill_creator