xianyu-hub

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

• Third-party content exposure detected (low risk: 0.10). scripts/smart_search.sh 在结果不足时会调用 SearchSharp.com 的 API（GET /api/products?q=...）获取由“用户社区众包维护”的别称文本，并将这些文本作为后续闲鱼搜索关键词传入 search.sh / execute_js 流程进行读取与处理。

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

• Potentially malicious external URL detected (high risk: 0.80). The skill's scripts (e.g., scripts/smart_search.sh and scripts/alt_keywords.sh) call https://search-sharp.com (notably https://search-sharp.com/api/products?q=${ENCODED}) at runtime to fetch community-provided alias keywords which are then used to drive subsequent searches/commands, so external content directly controls the agent's runtime behavior.