The Agent Skills Directory

[SAFE]: The skill serves as a technical manual and documentation set for the shadcn/ui framework, containing no executable malicious logic or instructions to bypass safety guidelines.
[INDIRECT_PROMPT_INJECTION]: The guide documents the framework's capability to fetch and install components from external registries (e.g., via npx shadcn add @namespace/component). This describes an inherent supply chain risk within the tool's ecosystem rather than a vulnerability in the skill itself. The skill provides necessary safety warnings for this feature.
Ingestion points: Registry URLs provided to the CLI or configured in the components.json configuration file.
Boundary markers: The documentation explicitly warns: 'Community registries are maintained by third-party developers. Always review code on installation.'
Capability inventory: The documented shadcn CLI is capable of file system writes, project configuration modification, and npm package installation.
Sanitization: The framework's core philosophy is 'Open Code,' meaning all generated code is added to the user's project for manual review and modification.

shadcn-guide