The Agent Skills Directory

[PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface as it ingests untrusted user text and file contents for processing.
Ingestion points: The skill reads prose directly from user chat or from file paths provided by the user in Step 1.
Boundary markers: There are no explicit delimiters or instructions to the agent to treat the input text strictly as data rather than instructions.
Capability inventory: The skill has access to powerful tools including Read, Write, Edit, and Skill (allowing it to invoke other agents/skills).
Sanitization: The input is passed to sub-skills without sanitization or escaping.
[SAFE]: The skill's primary function is to chain internal sub-skills (superpower-writing:humanizer and superpower-writing:writing-clearly-and-concisely) to perform linguistic analysis and editing. It uses standard file-system operations and follows modular design patterns without evidence of malicious intent or exfiltration.

polish