tdesign-miniprogram

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

• Third-party content exposure detected (high risk: 0.80). The skill's chat components (e.g., t-chat-content / t-chat-markdown, t-attachments, and t-chat-sender) render and preview arbitrary message markdown, attachments and remote URLs (examples: items with url fields like https://example.com/... and markdown links), which means untrusted user- or web-provided content is ingested and displayed as part of the agent's runtime workflow.