The Agent Skills Directory

[COMMAND_EXECUTION]: The skill includes instructions to run basic shell commands like ls, cat, and grep. These are used to inspect the local project structure for existing image assets, brand directories, and theme configuration files (e.g., tailwind.config.js). This process allows the agent to generate designs that are contextually aware and consistent with the user's existing project.
[SAFE]: No malicious patterns such as data exfiltration, network requests to untrusted domains, or prompt injection attempts were identified. The shell commands are limited to read-only discovery of non-sensitive project files and do not involve privilege escalation or persistence mechanisms.

design-canvas