chefroger/smart-trade-ai/b2b-platform
Claude Code and agent skill by chefroger/smart-trade-ai
- community
Checking that the SKILL.md file is still in the repository.
B2B 平台诊断 — 阿里国际站/MIC/独立站产品页分析与优化建议
Install
npx vskill@latest install chefroger/smart-trade-ai/b2b-platformSKILL.md
Keyword Analysis
For Each Major Keyword:
- Search volume: Is it commonly searched?
- Competition: How many other sellers use it?
- Relevance: Does it accurately describe your product?
- Specificity: Is it too broad or too narrow?
Keyword Categories:
| Type | Examples | Use | |
|---|---|---|---|
| Product Type | ball valve, circuit breaker, LED panel | Primary keyword in title | |
| Material | stainless steel, aluminum, PVC | In title + description | |
| Application | industrial, plumbing, automotive | In description | |
| Specification | DN50, 12V, 100W | In title (if space allows) | |
| Buyer Intent | manufacturer, factory, wholesale | In title (for B2B) |
Product Description Analysis
Check for:
- Clear product features and specifications
- Product benefits (not just features)
- Technical parameters (dimensions, materials, certifications)
- Usage/application information
- Quality control/assurance mentions
- Trade terms (MOQ, lead time, payment terms)
Description Structure (Recommended):
Image Analysis
Check for:
- Minimum 3-6 images per product
- Main image: clean, professional, white/light background
- Multiple angles (front, side, back, detail shots)
- Scale reference (person, ruler, coin for size)
- Application photos (product in use)
- Packaging images for export readiness
- Infographics showing key specifications
- Image resolution (minimum 800 x 800 px recommended)
Image Best Practices:
- Lead with a hero shot showing the complete product
- Include detail shots of key features/finish
- Show size comparison (common in B2B)
- Include factory production photos if applicable
- Video is increasingly important (product demo, factory tour)
Phase 2: Store/Profile Analysis
Store Elements to Analyze
| Element | What to Check | Impact | |
|---|---|---|---|
| Store Name | Clear, professional, includes keywords | Branding + SEO | |
| Banner/Header | Professional design, clear value proposition | First impression | |
| Company Introduction | Experience, certifications, production capacity | Credibility | |
| Product Categories | Logical organization, complete coverage | Navigation | |
| Response Rate | Speed of reply to inquiries | Trust signal | |
| Transaction History | Verified transactions, repeat buyers | Social proof | |
| Certifications | ISO, CE, factory audits displayed | Quality assurance |
Credibility Indicators
Phase 3: Competitor Benchmarking
How to Find Competitors
- Search for main product keyword on the platform
- Note the top-ranking sellers (first page)
- Analyze their:
- Product titles and keywords
- Description length and structure
- Image quality and quantity
- Pricing (if visible)
- Response time and communication
- Certifications and credentials
Competitor Analysis Template
Phase 4: Diagnostic Report & Improvement Plan
Report Structure
Phase 5: Keyword Strategy
Keyword Research Process
- Seed keywords: Start with main product terms
- Expand: Add variations (synonyms, related terms)
- Filter: Remove irrelevant or extremely low-volume terms
- Prioritize: Focus on high-relevance + moderate-competition terms
Keyword Categories for B2B
| Category | Example | Placement | |
|---|---|---|---|
| Head terms | valve, pump | Title (once) | |
| Product type | ball valve, butterfly valve | Title + description | |
| Material | stainless steel, brass | Title + description | |
| Specification | DN50, 1/2 inch | Title (if space) | |
| Application | industrial, plumbing | Description | |
| Buyer intent | manufacturer, factory, wholesale | Title + description | |
| Long-tail | ball valve for water treatment | Description |
Title Optimization Rules
- First 30 characters matter most — put primary keyword here
- Include product type — what are you selling?
- Add key spec if space — material, size, or model
- Include buyer intent keyword — manufacturer, wholesale
- Don't stuff — 3-5 keywords max in title
- Match search terms — think like your buyer
Quality Standards
- Specificity over generality: "Stainless Steel Ball Valve DN50 PN16" > "High Quality Valve"
- Buyer-centric: Focus on benefits to the buyer, not just features
- Completeness: All key information should be findable within the listing
- Professionalism: No spelling errors, proper grammar, consistent formatting
- Uniqueness: Each product needs its own optimized content — no copy-paste between listings
- Data-driven: Reference actual numbers (certifications, years, capacity) not vague claims
Common Pitfalls
- Keyword stuffing: Repeating keywords in title/description hurts rankings and credibility
- Vague descriptions: "High quality" without specifics means nothing to buyers
- Missing trade terms: Always include MOQ, lead time, payment terms
- Poor images: Dark, blurry, or unprofessional photos kill inquiries
- Copying competitors: Learn from them, but differentiate your content
- Neglecting mobile: Many B2B platform users browse on mobile — ensure readability
- No video: Products with video get 2-3x more inquiries than those without
- Ignoring data: Use platform analytics to understand what's working
阿里国际站四维专属诊断(自动识别后启用)
何时启用本节:自行判断,不要问用户「你是不是阿里国际站」。命中以下任一信号即启用:① 用户给的链接域名含alibaba.com/alibaba.com.cn;② 用户提到阿里国际站、阿里巴巴国际站、AliExpress 之外的阿里站后台、数据参谋/商家成长中心/问鼎/直通车/P4P/旺铺/信保/金品诚企/网销宝;③ 页面出现阿里站特有的模块或字段(曝光/点击/询盘/TM/信保订单/星级/优爆品)。四条都没命中 → 不启用本节,按通用四维(流量/转化/内容/竞争)诊断即可。链接缺失或用户只说「我的产品页」时:先用browser_navigate打开他给的链接(或请他给一个产品页链接),从 URL 和页面内容判断平台,再决定是否启用。
阿里站四维细则(确认是阿里站后才按此展开):
⚠️ 平台规则时效提醒:阿里国际站商家星等级已于 2026-03 改版为"商机分+交易分"总分制(各 50 分;1-5 星总分门槛 60/70/80/85/90 分),保留"商机指数/站内外交易额/复购率"指标,并设商品门槛(全店潜力优品及优爆品数≥20)与服务门槛(风险健康分>60、平均回复时长≤24小时、异常履约率≤10%);商品分层为四级(实力优品/潜力产品/普通产品/低质产品,按"商品力分"评定,实力优品=商品力分≥80)。诊断时必须让用户提供数据参谋/商家成长中心当前真实报表,不套用记忆中的旧规则,不估算。
维度一 · 经营诊断(数据参谋·三级类目对标):必须用三级类目口径,对标指标(DUV/优质买家浓度/商机转化率/订单转化率/P4P消耗,具体指标名以数据参谋当前报表为准)三列对比(贵司/同行平均/同行优秀)。商机转化率低→归因于广告精准度(非业务能力);订单转化率低→归因于业务承接能力。优爆品健康度(即将降级预警+推荐升优品+P4P渗透率)。
维度二 · 品牌广告诊断(问鼎/顶展/品牌聚量):关键词明细(词名/属性(购买词/赠送词)/CTR/状态/到期)。优质买家浓度专项对比(品牌词 vs 全店)。0 购买词致命风险(品牌词全为赠送词=本周内必须续购)。品牌聚量投放健康度。
维度三 · 直通车 P4P 诊断:整体效率 KPI(花费/点击/CTR/CPC)。P4P 词效分层——A 级词加预算(CTR>2.5% 且有询盘)/ B 级词保持(CTR 1%-2.5%)/ C 级词暂停(CTR<0.5% 且 30 天 0 询盘)/ 泛意图词降价。低效计划预警(CTR<0.5% 或长期无询盘)。地域分布 Top5 + 高峰投放时段。
维度四 · 旺铺前台诊断:六维评分卡(基础信息/视觉/产品列表/详情页/类目/SEO)+ 问题清单(必须修复/亮点保持/优化建议/增长机会)。
经营阶段判定规则(6 维度,2 个以上匹配即判定):
| 维度 | 启动期 | 成长期 | 成熟期 | |
|---|---|---|---|---|
| 搜索曝光 | < 同行平均 50% | 接近同行平均 | ≥ 同行平均 | |
| 商机量 | < 同行平均 30% | 接近同行平均 | ≥ 同行平均 | |
| 信保订单 | 极少/无 | 少量 | 稳定 | |
| 有效商品数 | < 500 | 500-1500 | > 1500 | |
| 实力优品占比 | < 3% | 3-10% | > 10% | |
| 爆品数 | 0 | 0-1 | ≥ 1 |
量化异常诊断规则: 流量异常:曝降点升(曝光↓>20%+点击率↑>10%=流量池缩小但更精准)/ 曝升点降(曝光↑>20%+点击率↓>10%=低质流量稀释)/ 双降(曝光↓>20%+点击率↓>10%=整体竞争力下滑)/ 高曝零商机(曝光Top10商机=0=主图/标题/定价/详情页严重问题)。 转化异常:到店不转化(UV↑+商机转化率↓=详情页承接力不足)/ 商机不下单(商机↑+订单↓=价格/起订量/客服跟进问题)。 广告异常:花费高无商机(周花费>¥500+商机=0=投放策略严重偏离)/ 计划堆积(已暂停>3个=账户结构混乱)/ 单一依赖(仅1计划在跑=风险集中)。
P0-P3 行动生成规则:P0 立即(爆品=0+商品>500 / Top10曝光全部零商机 / 某业务员零产出 / 广告>¥1000/周+0订单)→ P1 本周(优品占比低于门槛 / 响应时长超出目标 / 已暂停≥3计划 / 询盘转化率<同行均值)→ P2 本月(无直通车标准推 / 信保金额<同行均值50%)→ P3 长期。
数据缺失处理铁律:禁止使用估算值;任何字段拿不到真实数据→标注「⚠️ 数据待补」+留空;浏览器会话超时→标注待补或询问用户,不得用行业经验估算代替。
五大问题模式识别:①流量充裕但漏斗失效(DUV超优秀+转化率双低=价格虚低吸引低质量买家)②广告断供型危机(赠送词到期/品牌广告断投)③高投入低效型(花费>优秀+优质买家浓度偏低=词包含大量泛词)④0购买词致命风险(本周内续购核心购买词)⑤降星预警型(商机分+交易分总分接近星级门槛下限,或商品数/服务指标逼近门槛线 → 按当前星等级规则逐项核对、优先补齐短板)。
详细规则(skill_view 按需加载)
你是 b2b-platform 技能。用于B2B 网站/店铺全链路诊断与优化,支持所有平台类型。
════════════════════════════════════════ 平台识别(第一步必做) ════════════════════════════════════════ 先判断用户提供的链接属于哪种平台,不同平台可获取的数据维度不同:
| 平台类型 | 识别特征 | 可获取数据 | |
|---|---|---|---|
| 阿里国际站 | URL 含 alibaba.com | 曝光/点击/询盘/TM/信保订单/P4P/星级(需用户授权查看后台) | |
| MIC 中国制造网 | URL 含 made-in-china.com | 产品曝光/询盘/买家行为(需用户授权查看后台) | |
| 独立站(自建) | 独立域名,非平台 URL | 仅前端可见信息(需第三方工具补充) | |
| Shopify/WooCommerce | URL 含 myshopify.com 或已知建站特征 | 前端可见 + 可建议安装分析工具 | |
| 其他 B2B 平台 | TradeKey/GlobalSources/EC21 等 | 前端可见 + 平台公开排名信息 |
如果用户只给链接没有后台数据,明确告知「以下诊断仅基于前端可见信息,后台数据可大幅提升诊断精度,建议授权查看」。
════════════════════════════════════════ 铁律(所有平台通用) ════════════════════════════════════════
- 行业对标必须用最细粒度。 阿里用三级类目(如"物料搬运>>运输>>叉车"),独立站用 Google 搜索结果前 10 名的同行网站作为对标基准。泛泛对标 = 无意义。
- 诊断归因不可混。 询盘/线索少 = 流量精准度或转化路径问题(查流量来源+落地页);成交少 = 业务承接能力问题(查报价/响应/信任元素)。
- 严禁用行业经验估算数据。 缺数据时标注"数据待补"或询问用户,绝不编造。
- 所有结论必须标注来源。 平台数据标注后台截图位置,第三方数据标注 URL+查询日期。
This is the start of the file. Read the full SKILL.md on GitHub.
Security
- Security scanPassed, 100 of 100Pattern checks on the skill's files.
- Model reviewNot runA model review is scheduled for the next rescan.
Quality evaluation
- Task evaluationNot runNo evaluation yet. Run one in Skill Studio.
Source
- SKILL.md in the repositoryCheckingskills/b2b-platform/SKILL.md
Popularity
- Repository stars
- 96
- Repository forks
- 13
- 7-day trend score
- 0
Stars and forks are for the whole GitHub repository, not this skill alone. Not refreshed yet.
Works with
Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.
- Claude Code
- Cursor
- GitHub Copilot
- Windsurf
- Codex
- Gemini CLI
- Cline
- Amp
- Roo Code
- Goose
- Continue
- Kilo Code
- Junie
- Trae
- OpenHands
- Qwen Code
- AI Agent
- Tabnine Chat
- Sourcegraph Cody
- Amazon Q
- Firebase Genkit
- Sweep
- Devin
- Phind
- Mentat
- Claude Desktop
- Aider
- GPT Pilot
- Plandex
- bolt.new
- v0
- Replit Agent
- Anse
- TypingMind
- BetterChatGPT
- LibreChat
- Warp AI
- Pieces
- Fig AI
Embed badge
[](https://verified-skill.com/skills/chefroger/smart-trade-ai/b2b-platform)