minimax-h3

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

• Third-party content exposure detected (medium risk: 0.30). At runtime, outsider-authored free text is ingested directly as --prompt (and optional --system) by the CLI and forwarded in the messages body to POST /openapi/v1/chat/text or POST /openapi/v1/generation/video via skill_generate() → cmd_video() → client.generate_video() / client.chat_text(), with no selection step.