The Agent Skills Directory

[SAFE]: No malicious behavior or security vulnerabilities were detected. The skill is designed for linguistic editing and copywriting tasks.
[PROMPT_INJECTION]: The instructions are focused on linguistic style and tone. There are no attempts to override agent safety protocols, extract system prompts, or bypass constraints.
[DATA_EXPOSURE]: The skill searches for a local file named brand-brief.md to determine brand-specific tone and vocabulary. This is a standard configuration pattern for personalization and does not involve accessing sensitive system credentials or exfiltrating data.
[EXTERNAL_DOWNLOADS]: The skill does not contain any network operations or external URL references for downloading code or content.
[REMOTE_CODE_EXECUTION]: There are no scripts or commands that execute code, install packages, or spawn subprocesses.

humanizer-de