shadcn

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly instructs the agent to fetch and read external documentation, registry items, and arbitrary URLs (e.g., via npx shadcn@latest docs <component>, npx shadcn@latest add with registry/URL inputs, and the MCP view/get_item_examples_from_registries tools), which are untrusted community/third‑party sources whose content the agent must interpret to decide installs, diffs, and merge actions—creating a clear avenue for indirect prompt injection.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly instructs fetching remote component docs and raw example files at runtime (e.g., https://raw.githubusercontent.com/.../examples/button-example.tsx), which would be retrieved and injected into the agent's context to guide its instructions and component installation behavior.