shadcn/ui Component Library

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (medium risk: 0.65). The required runtime workflow is the shadcn CLI/registry/MCP flow (e.g., npx shadcn@latest add / mcp init / registry URLs), which can fetch and ingest component source text from external registries or public sources into the agent/LLM context, creating an indirect prompt-injection surface.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The skill configuration explicitly runs remote code at runtime via npx (e.g., "npx shadcn@latest mcp") and also declares registries that are fetched at runtime (e.g., "https://acme.com/r/{name}.json" and "https://internal.company.com/{name}.json"), so external content is retrieved and can install/execute code or control what is added to the project.