alphaear-signal-tracker

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly instructs agents to fetch and ingest open/public web and social-media content (e.g., NewsToolkit.fetch_hot_news listing weibo/zhihu/douyin/xueqiu, NewsToolkit.fetch_news_content which uses ContentExtractor.extract_with_jina to fetch arbitrary URLs, and SearchToolkit.web_search) and the FinResearcher/Trend prompts require the agent to read and use that third‑party content in analysis, which clearly exposes it to untrusted user-generated sources that could carry indirect prompt injection.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The skill's ContentExtractor calls the Jina Reader API at https://r.jina.ai/ at runtime to fetch webpage content (via fetch_news_content) which is then injected into agent prompts/research context and thus can directly influence model instructions and outputs.