The Agent Skills Directory

[SAFE]: The skill performs legitimate text processing tasks as described. It uses file system tools (Read, Edit, Write) exclusively for reading input and saving rewritten prose back to the original file path as requested by the user.
[INDIRECT_PROMPT_INJECTION]: The skill has a standard attack surface for indirect prompt injection as it processes untrusted data from files or user input and possesses the capability to write back to the filesystem. However, its specific task of structural rewriting and removal of AI-isms significantly mitigates the risk of executing embedded instructions.
Ingestion points: The skill reads external file content using the Read tool as defined in SKILL.md.
Boundary markers: No explicit boundary markers or 'ignore embedded instructions' directives are defined for the processed content.
Capability inventory: The skill utilizes Write and Edit tools to modify files on the local disk (SKILL.md).
Sanitization: No specific sanitization or escaping of the input text is performed before processing.

humanize