The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes dynamic context injection (using the !command syntax) to execute shell commands at load time. These commands (ls and cat) are used to detect the presence of project-specific files like DEVICE.md and PROJECT_BRIEF.md. This is a legitimate use of the feature for state detection and does not involve sensitive file access or network activity.
[SAFE]: The skill's primary purpose is routing. It restricts its analysis to local project markers and provides official documentation links to the vendor's (Luxonis) website. No data exfiltration, obfuscation, or malicious prompt injection patterns were detected.

luxonis-guide