The Agent Skills Directory

[SAFE]: The skill consists entirely of markdown documentation and static code samples for frontend developers. No executable scripts, network exfiltration logic, or privilege escalation patterns were detected.
[EXTERNAL_DOWNLOADS]: The skill references several well-known libraries in its code examples (e.g., @dnd-kit/core, react-hook-form, zod, sonner) and provides links to official documentation (W3C, Nielsen Norman Group, etc.). These are purely informational and do not involve untrusted remote code execution.
[PROMPT_INJECTION]: No attempts to override agent behavior or bypass safety guidelines were found. The 'Persuasion Ethics' section documents malicious UI patterns (dark patterns) specifically to instruct the agent to avoid them.

interaction-patterns