react-shadcn

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

• Third-party content exposure detected (high risk: 0.80). The SKILL.md Agent Workflow explicitly requires using a research agent to "verify latest shadcn/ui docs via Context7/Exa" and MCP tools (mcp__shadcn__search_items_in_registries, view_items_in_registries, get_item_examples_from_registries) to query public shadcn registries—clear, mandatory fetching and interpretation of open/public third‑party content that will influence implementation decisions and tool use, enabling indirect prompt injection risk.