The Agent Skills Directory

[SAFE]: The skill functions as an instructional framework for UX design sessions, utilizing structured prompts and confirmation gates to manage workflow state.
[SAFE]: File system activity is limited to creating markdown reports in a local documentation directory, which is consistent with the stated purpose of facilitating a design workshop.
[PROMPT_INJECTION]: The skill ingests untrusted data from the user during the 'Problem Definition' phase (Fase 1) in SKILL.md. It lacks boundary markers to delimit user input from instructions and does not perform input sanitization. This creates an indirect prompt injection surface; however, the risk is low as the skill's capabilities are restricted to generating text and writing local documentation files.

design-studio