gradio

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

• Third-party content exposure detected (high risk: 0.80). The skill's documentation and examples include components that ingest and display untrusted third-party content—e.g., gr.Chatbot (displays user-submitted messages) and gr.HTML examples that load external resources (https://upload.wikimedia.org/...) and run js_on_load that can set props/trigger events—allowing remote or user-generated content to be interpreted and to trigger Gradio events that can change app behavior.