agent-ui

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

• Third-party content exposure detected (medium risk: 0.65). The required workflow references a public, runtime-fetched agent component/config from ui.inference.sh (via npx shadcn@latest add https://ui.inference.sh/r/agent.json in SKILL.md), which can cause outsider-authored free-form text to be ingested into the agent/LLM context.