problem-doc-model-selector

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

• Third-party content exposure detected (high risk: 0.90). The skill's mandatory "外部数据检查 (必做)" requires generating data_requirements.json with explicit url entries "以便 authoritative-data-harvester 自动获取", and the "Memory Interaction (必做)" step requires checking external literature/data indexes (e.g., g-sci) and incorporating them into model selection, which indicates the agent will ingest and act on public third‑party data that can influence its decisions.