The Agent Skills Directory

Command Execution (SAFE): The skill instructs the agent to use npx shadcn@latest for project initialization and component installation. This is the standard and recommended way to interact with the shadcn/ui ecosystem and is necessary for the skill's primary purpose.
Indirect Prompt Injection (LOW): The skill ingests untrusted user input (queries for components or features) and interpolates them into tool calls.
Ingestion points: User-provided search queries in SKILL.md (e.g., 'add a date picker', 'build a login form').
Boundary markers: No specific delimiters or safety instructions are used when passing the query variable to shadcn___search_items_in_registries or shadcn___get_item_examples_from_registries.
Capability inventory: The skill can recommend command-line executions (npx shadcn) and file modifications (frontend design customization).
Sanitization: No explicit sanitization of user-provided strings is performed before tool invocation.

shadcn-management