The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses dynamic context injection in SKILL.md to execute npx shadcn@latest info --json at load time. This allows the agent to automatically synchronize its context with the user's project configuration, including Tailwind versions, import aliases, and installed components.
[EXTERNAL_DOWNLOADS]: The skill facilitates the installation of UI components and the retrieval of documentation/examples from external sources using the shadcn CLI tool. This is a standard workflow for the shadcn/ui ecosystem.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection due to its reliance on external and local project data.
Ingestion points: Project metadata is ingested from npx shadcn@latest info and component documentation/examples are fetched from external URLs via npx shadcn@latest docs.
Boundary markers: No explicit delimiters or instructions to ignore embedded commands are present when processing the fetched content.
Capability inventory: The skill has the capability to execute shell commands (npx), write files to the project directory (add), and perform network operations.
Sanitization: There is no evidence of sanitization or validation performed on the data fetched from documentation URLs or local configuration files before it is processed by the agent.

shadcn