The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses bash and grep commands for legitimate project scanning tasks such as directory creation, listing files, and searching for UI component patterns (e.g., CVA variant detection).
[EXTERNAL_DOWNLOADS]: The skill references configuration and documentation from the ${CLAUDE_PLUGIN_ROOT} path, which is part of its own internal framework environment and does not involve untrusted remote sources.
[SAFE]: The skill implements a strict 'Hard Stop' protocol, requiring explicit user approval via AskUserQuestion before generating any plans or writing files to the disk.
[SAFE]: File writing is constrained to the $JAAN_OUTPUTS_DIR/frontend/** directory, preventing arbitrary writes to sensitive system or project areas.

frontend-story-generate