协作头脑风暴
brainstorming
在实现前澄清需求、探索设计选择,并形成清晰的开发方向。
sealeap-tianlu-amazon-order-drop-multifactor-diagnosis【开发与工程】When order volume drops during a period with multiple concurrent changes (an external ad-channel disruption, a tightened compliance review cycle, and warehouse or inbound congestion), evaluate each candidate cause against its own account-specific evidence — external traffic-source records, category-specific compliance notices actually received, and F…
独立 GitHub 快照 · 自动分类 · 以原作者文档为准
When order volume drops during a period with multiple concurrent changes (an external ad-channel disruption, a tightened compliance review cycle, and warehouse or inbound congestion), evaluate each candidate cause against its own account-specific evidence — external traffic-source records, category-specific compliance notices actually received, and FBA inbound or receiving status — before assigning weight to any single factor. Produces a per-factor evidence table rather than a single narrative cause.
用户未指定时采用“诊断”。
缺失项必须标为 NEEDS_EVIDENCE;不得猜数字、补属性或把不同站点、ASIN、变体、币种和时间窗混在一起。
先读取 references/playbook.md,确认该方法适用于当前对象。按以下顺序执行:
最后做数据充分性检查,并把结论分成 FACT / ESTIMATE / HYPOTHESIS / UNKNOWN。若关键证据不足,状态写 HOLD。
仅在自有数据不足且当前任务确实需要外部证据时,读取 references/mcp-data-plan.md,再使用 scripts/mcp_research.py。本 Skill 的外部取数目的:如需判断下滑是账户自身问题还是同类目大盘现象,可用第三方类目趋势数据连接器核对同期大盘走势作为参照证据,不单凭主观归因或个案传闻。
doctor,再 search-tools 和 describe;工具名及参数以实时 tools/list 与 inputSchema 为准。tools/call 或 Actor 可能计费;先展示 Provider、工具、无密钥业务参数、预计成本与输出位置,核对已有授权覆盖后才加 --allow-cost;该标志不是费用上限。方案状态使用 READY FOR REVIEW / DRAFT / HOLD / STOP;如已执行,另行记录实际结果及回读证据。未得到明确批准时,不得声称已修改线上对象。