The Agent Skills Directory

[SAFE]: The skill is a standard prompt-based utility for structured writing. It does not employ any suspicious tools, network operations, or persistence mechanisms.
[PROMPT_INJECTION]: The skill processes untrusted input from markdown files, which constitutes a potential indirect prompt injection surface. The agent is instructed to paraphrase and structure this material, which provides a layer of interpretation between the input and output. * Ingestion points: Reads raw material from an input markdown file as described in SKILL.md. * Boundary markers: The instructions do not define specific delimiters for the input content. * Capability inventory: The skill utilizes file read and write tools to manage the article drafting process. * Sanitization: No explicit sanitization or filtering of the input text is performed.

writing-shape