options
Pass
Audited by Gen Agent Trust Hub on Aug 11, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is purely instructional, designed to change how the AI agent presents choices to the user (converting prose questions to multiple-choice tools).
- [EXTERNAL_DOWNLOADS]: The documentation provides
npxcommands for installation and update purposes. These commands point to the author's official GitHub repository (github.com/leoluyi/skills), which is consistent with the provided author metadata. As these are standard developer tools for skill management from a recognized vendor source, they represent intended functionality rather than a security risk. - [COMMAND_EXECUTION]: While the skill instructions refer to using a 'tappable-choice tool', this refers to UI elements within the AI assistant's interface rather than executing system-level commands or shell scripts.
Audit Metadata