The Agent Skills Directory

[SAFE]: No security issues detected. The skill focuses on defining a persona and execution flow for user interface design, including interaction protocols with a context manager and status updates. While the skill has access to tools like Bash and Edit, no malicious use of these capabilities is present in the instructions.- [SAFE]: The skill exhibits a potential surface for indirect prompt injection by ingesting design context from an external 'context-manager' agent. However, this is a standard pattern for collaborative agent systems and does not currently incorporate any dangerous interpolation or lack of boundaries that would elevate risk beyond the baseline for this type of architecture.

ui-designer