academic-search

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill explicitly asks the user to supply API keys and instructs the agent to write them verbatim into a .env (e.g., "写入 SEMANTIC_SCHOLAR_API_KEY=xxx"), which requires the LLM to handle and embed secret values directly, creating a clear exfiltration risk.

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

• Third-party content exposure detected (medium risk: 0.30). 在 scripts/academic_search.py 的运行路径中，用户输入作为 query 直接驱动调用外部检索 API（如 https://metaso.cn/api/v1/search 与 Semantic Scholar/Crossref/PubMed/arXiv 等），并把返回的自由文本字段（例如 metaso 的 snippet 等）拼入最终输出。