shadcn-component-review

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

• Third-party content exposure detected (high risk: 0.75). The required workflow can ingest outsider-authored free text when the user pastes or requests review of an existing component (e.g., “User says ‘I built this’ and pastes a component” / “Review this component”), and that pasted code/text is then analyzed by the agent’s LLM.