算法艺术创作
algorithmic-art
通过代码和可调参数创作生成艺术与交互式视觉效果。
sealeap-tianlu-amazon-ai-readable-title-structure【设计与创意】Restructure product titles from keyword-stuffed strings into a semantically ordered two-part layout (brand plus core category and attributes first, then differentiating function and use-case terms), then validate each candidate term's real search relevance through storefront search evidence and keyword-volume data before committing. Treats the premis…
独立 GitHub 快照 · 自动分类 · 以原作者文档为准
Restructure product titles from keyword-stuffed strings into a semantically ordered two-part layout (brand plus core category and attributes first, then differentiating function and use-case terms), then validate each candidate term's real search relevance through storefront search evidence and keyword-volume data before committing. Treats the premise that stuffed titles now carry an algorithmic penalty as an unverified hypothesis to test with the account's own indexing and conversion data, not an established rule.
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:补充候选标题词的搜索热度与相关词代理数据,用于核实结构化后的候选词是否为真实被检索的高频词,而非仅凭主观判断保留。
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。