Qwen

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

• Third-party content exposure detected (high risk: 0.80). The skill calls external Alibaba Cloud Model Studio endpoints (e.g., the https://dashscope.../models and /chat/completions URLs shown in SKILL.md and api-patterns.md) and uses those live model responses to drive routing and tool-calling behavior (see tool-calling.md), so untrusted third‑party content can influence agent actions and enable indirect prompt injection.