The Agent Skills Directory

[SAFE]: The skill focuses on marketing and product copywriting, utilizing a set of 'anti-AI-slop' rules and JSON output structures. Analysis of the instructions and implementation guides shows no evidence of prompt injection, data exfiltration, or unauthorized code execution.
[INDIRECT_PROMPT_INJECTION]: The skill processes content from a local project file (.moai/project/brand/brand-voice.md). While this ingestion point represents a potential surface for indirect injection, the skill lacks capabilities to perform high-risk actions (network requests, system modifications, or arbitrary file writes). Furthermore, the output is strictly constrained by a structured JSON schema and a mandatory 'anti-slop' checklist, effectively neutralizing the risk of instruction override via external project data.

design-copywriting