The Agent Skills Directory

[SAFE]: The skill provides instructions for a writing style editor. It focuses on linguistic adjustments (sentence length, paragraph structure, and word choice) to improve the readability and 'human feel' of Chinese text.
[SAFE]: It references an external skill humanizer-zh as an optional dependency for surface-level processing. This is a standard modular design and does not involve unauthorized remote code execution or data exfiltration.
[SAFE]: All referenced files are local documentation (references/check-criteria.md, references/output-template.md) containing guidelines and templates for the AI's internal reasoning process.
[SAFE]: No network operations, credential access, or sensitive file system interactions were found in the instructions or metadata.

humanize-plus