The Agent Skills Directory

[SAFE]: The skill implements a legitimate prompt engineering technique described as 'Prompt Repetition'. The logic focuses on duplicating strings to enhance attention mechanisms in causal language models.
[SAFE]: The provided Python script prompt_repetition_transformer.py uses standard string manipulation, regex-based task detection, and safe conditional logic. It does not utilize dangerous functions such as eval(), exec(), or external network requests.
[SAFE]: The skill contains no signs of obfuscation, credential harvesting, or unauthorized privilege escalation. The metadata and implementation are consistent with the stated purpose of optimization for lightweight models.
[SAFE]: Non-English (Korean) comments in the source code are descriptive in nature and do not contain hidden instructions or homoglyph attacks.

prompt-repetition