Turn an Amazon keyword universe into a clean taxonomy and campaign map using relevance, intent, roots, negatives, and evidence from authorized reports. Use when the user asks how to collect keywords, perform word-root analysis, choose auto versus broad versus exact targeting, or prevent broad campaigns from drifting. Produce a draft architecture and never apply ad changes without explicit approval.
npx skills add https://github.com/xjli360/sealeap-amazon-skills --skill sealeap-amazon-keyword-selection-and-campaign-mapping
把杂乱的关键词集合转成可执行的分类、否定和投放结构,使探索范围与高转化目标同时可控。
缺失的数据要明确列为缺口,并把结论标成事实、估算或假设;不要补造数字。
只有在本任务确实需要外部市场、竞品、关键词或公开网页证据时,才读取 references/mcp-data-plan.md,并使用 scripts/mcp_research.py。
统一大小写、单复数和常见拼写,保留原始来源字段,删除重复项但不丢失证据链。
至少区分高意图属性词根、覆盖型高频词根、通用词、品牌词、竞品词和不相关词根。
按事实相关性、购买意图、流量、竞争、预估转化和利润空间评分;缺失数据不伪造,降级为待验证。
自动用于受控发现,广泛或词组用于词根扩展,精准用于已验证词,商品投放用于相似详情页或类目机会。
明确哪些词做精准否定、哪些词根可做词组否定,并记录否定原因和复核人。
按固定窗口把出单搜索词迁移、把高耗无转化词降级或否定,并同步检查广告间的重复覆盖。
结尾列出数据窗口、关键假设、证据缺口、风险和下一步;如包含待执行动作,单独放在“待批准变更”中。
Take xjli360/sealeap-amazon-keyword-selection-and-campaign-mapping from the repository into ~/.claude/skills for personal
use, or into .claude/skills inside a project.
The agent identifies a skill by the name field in its header. Two skills with the
same name cannot sit side by side — one of them will be ignored.