stock-market-hotspot-discovery

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

• Third-party content exposure detected (medium risk: 0.65). 该技能在运行时通过 scripts/get_data.py 的 _http_call_hotspot_discovery() 向 HOTSPOT_DISCOVERY_URL 发起请求并将响应中的 data.displayData/content 等可读文本提取为 result["content"]，再作为 Markdown 输出/保存；该外部 API 的返回文本属于“公共/第三方运行时获取内容”，可能包含他方编写的自由文本从而进入 LLM 上下文。

HIGH W008: Secret detected in skill content (API keys, tokens, passwords).

• Secret detected (high risk: 1.00). I found a hard-coded, high-entropy API key default in the script: EM_API_KEY is set to a literal value ("em_fjFqd4YB6Cqs52LF48XWbMDdLNq6MyNg") and is sent in request headers ("em_api_key"). This is not a placeholder (e.g., YOUR_API_KEY) or a simple setup password — it appears to be a real API credential baked into code and thus qualifies as a secret. No other high-entropy credentials (private keys, tokens, or passwords) are present; other values (URL, paths, timeout) are non-sensitive configuration.