shadcn

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

• Third-party content exposure detected (high risk: 0.90). The skill's workflow (SKILL.md) and MCP tool docs (mcp.md/cli.md) explicitly require running commands like npx shadcn@latest docs and using MCP view_items_in_registries/view to fetch and read component docs, registry JSON, and arbitrary URLs (including community registries and user-provided URLs), which are untrusted third‑party sources that the agent must interpret to decide installs/merges and subsequent actions.

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

• Potentially malicious external URL detected (high risk: 1.00). The skill explicitly fetches remote registries and component/docs files at runtime (e.g. community registry index https://ui.shadcn.com/r/registries.json and registry templates like https://acme.com/r/{name}.json) which are used to retrieve remote component source and installation instructions that directly determine what the CLI/agent installs into the project.