shadcn

CRITICAL E005: Suspicious download URL detected in skill instructions.

• Suspicious download URL detected (high risk: 0.70). Some URLs point to third-party/untrusted registry endpoints (api.npoint.io and custom registry URLs like acme.com/private.com) that can deliver arbitrary registry JSON used by the CLI to inject source files into a project, which is a potential supply‑chain/malware vector; official ui.shadcn.com and raw.githubusercontent.com links are benign in this context.

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

• Third-party content exposure detected (high risk: 0.72). The required workflow includes fetching component documentation/examples and viewing registry items (e.g., npx shadcn@latest docs then fetch URLs, and npx shadcn@latest view/add --view), which can ingest outsider-authored free text/code from public registries/GitHub/URLs into the agent’s LLM context.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly instructs the agent to run the CLI and fetch external documentation/registry files at runtime (e.g. https://ui.shadcn.com/init?base=radix&style=nova&...), and those fetched remote docs/registry files would be used to drive the agent's component-generation instructions.