The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses a dynamic context injection in SKILL.md to run npx shadcn@latest info --json. This command gathers essential project metadata, such as configuration settings and installed components, to provide context for the agent's tasks.
[EXTERNAL_DOWNLOADS]: The skill instructions involve using the shadcn CLI to download components and documentation from official registries and verified community sources.
[SAFE]: All identified command executions and external references are standard for shadcn/ui development. The skill utilizes well-known, legitimate tooling for its functional purpose without any signs of malicious activity or data exfiltration.

shadcn