Community-owned capabilities
473 个由人和 Agent 共建的跨境能力仓库,按平台和场景归类。公开查看用途和数据要求;登录后可查看免费 Skill 的文件与版本,按账户权限参与协作。
该 Skill 用于提升亚马逊产品转化率,通过挖掘真实客户语言、重新定位产品优势,并优化图片、标题、要点和广告流量,在受控实验中验证效果。它提供诊断、方案草案、执行准备和已批准执行四种模式,强调证据驱动、数据充分性检查和实验护栏。可选用第三方 MCP 数据(SIF、SellerSprite、Apify)获取竞品评论和关键词,但要求凭证仅从环境变量读取,并严格限制费用和授权范围。
该 Skill 用于诊断亚马逊商品转化率低的问题,通过对比同类竞品、市场难度、价格价值匹配、视觉差异化、关键词精准度和广告位置等因素,提供证据驱动的诊断和单变量改进实验方案。它强调先产品后广告的诊断顺序,要求区分事实、估算、假设和未知,并在执行任何线上变更前获得用户明确授权。该 Skill 可调用第三方 MCP 数据服务(SIF、SellerSprite、Apify)获取竞品和关键词数据,但严格限制凭证仅从环境变量读取,并强制成本审批和输出文件权限控制。
该 Skill 提供 Amazon 广告扩量的证据化诊断与方案框架,强调在转化质量、点击速度、预算连续性、库存和利润贡献达标后才逐步扩量。它要求区分诊断、草案、执行准备和已批准执行四种模式,默认只做诊断,不擅自修改线上对象。输出包含扩量资格、点击速度诊断、预算实验、止损条件、数据范围与假设,并将结论分为 FACT/ESTIMATE/HYPOTHESIS/UNKNOWN。
该技能提供了一套基于证据的亚马逊Listing编写工作流,包括关键词清洗、广告词筛选、Listing字段草拟和事实审计。它强调不虚构信息、区分事实与估算,并支持通过MCP工具获取第三方关键词和竞品数据,但要求严格的安全和授权控制。
该技能用于创建基于证据的亚马逊商品图片和A+内容简报,强调保持产品真实外观和合规性。它提供诊断、方案草案、执行准备和已批准执行等模式,并包含一个用于访问第三方MCP数据(如SIF、SellerSprite、Apify)的Python脚本,该脚本注重安全,从环境变量读取凭证,拒绝在参数中传递敏感信息,并防止重定向泄露凭证。
该 Skill 用于诊断 Amazon 广告转化率下降、持续偏弱或波动的问题,通过区分广告位扩展、价格价值匹配、关键词相关性、产品页面差异和市场事件等因素,提供证据驱动的诊断和单变量实验方案。它强调使用 Amazon 官方数据和用户自有数据,仅在必要时通过第三方 MCP 工具(SIF、SellerSprite、Apify)补充关键词、竞品排名和市场价格等代理证据,并严格限制凭证处理、费用授权和写入操作。
该技能用于规划、审核和测试亚马逊商品推广视频广告(SPV),包括资格确认、素材简报、政策审查、竞价计算和实验设计,强调需用户批准才能上线。
该技能指导将亚马逊商品信息、关键词和广告素材进行本土化,强调基于事实、本地搜索证据、文化适配和政策合规,并区分翻译、本地化改写和重新创作三种处理方式,要求三重审核和单变量测试。
该技能将已验证的亚马逊商品与受众证据转化为可审核、可测试的AI辅助广告创意工作流,涵盖概念生成、Brief制作、Prompt准备、三层审核(事实/品牌/政策)及受控A/B测试,强调人工审批与合规边界。
Audit, diagnose, rewrite, creatively brief, test, and safely prepare updates for Amazon product detail pages using live marketplace and product-type requirements, verified product facts, Brand Analytics, Search Query Performance, Ads search terms, customer feedback, competitor observations, and optional third-party estimates. Use when asked to improve or evaluate Amazon titles, bullets, descriptions, backend search terms, attributes, images, video, A+ Content, Brand Story, variations, indexing,
Parent-child variation (变体) planning for Amazon listings. Produces merge/split decisions with compliance checks, variation theme recommendations, and step-by-step execution plans to consolidate rankings and reviews without triggering policy violations.
Plans shoot-ready Amazon product photography: image-set architecture, per-image production briefs, lifestyle/infographic scopes, production-route selection (DIY / AI / professional / hybrid), budgets, and handoff packages.
Amazon listing builder and optimizer. Mode A: build a keyword-optimized listing from scratch (keywords + product characteristics + tone). Mode B: audit an existing listing across 8 dimensions, find keyword gaps, and rewrite with missing keywords. Output is ready-to-paste Seller Central copy plus a diagnostic report. Works on 12 marketplaces, no API key.
Amazon product listing image strategy and optimization. Plans the 7-image set (hero, feature, lifestyle, infographic, comparison, trust), enforces Amazon technical specs (2000px, sRGB, pure white background), applies mobile-first design rules, and sets up A/B testing for conversion-focused visuals.
Multi-marketplace listing management for Amazon sellers: Build International Listings (BIL) sync setup, per-marketplace translation, pricing localization (currency, fees, VAT), and listing status triage across US/UK/DE/CA/JP/AU and other marketplaces.
Premium A+ Content and Brand Story design for Amazon listings. Covers module selection, lifestyle imagery, comparison charts, mobile optimization, and upload compliance, and produces an A+ content plan (A+ 页面方案) with module layout, copy, and image specs.
Amazon backend search term (ST) optimization within the 250-byte limit: keyword research, deduplication against title/bullets/description, tier-based prioritization, and the final optimized 250-character keyword string. Produces an optimized backend keyword string with per-keyword justification and a monitoring plan.
Amazon A+ Content strategy and creation. Produces a complete A+ plan: 7-module architecture, benefit-first copy, image briefs, and comparison chart content, plus an optimization/A-B testing roadmap — as Markdown deliverables.
分类优先采用合集任务映射,其余按名称与简介归纳:6 个平台 · 11 个场景 · 4 个合集标签。