The Agent Skills Directory

[SAFE]: The skill incorporates explicit security boundaries, specifically instructing the agent to mock backend behaviors and avoid connections to production databases, authentication, or payment providers.
[SAFE]: All file creation is scoped to a designated directory (docs/assets/prototype/), preventing arbitrary file system modification and maintaining project organization.
[SAFE]: The instructions emphasize using provided project context as evidence rather than absolute authority and include a verification step to ensure prototype integrity without executing untrusted code.
[SAFE]: No patterns of prompt injection, data exfiltration, or persistence mechanisms were detected in the skill instructions or associated reference files.

to-prototype