← Back to Skills Marketplace
329
Downloads
1
Stars
1
Active Installs
3
Versions
Install in OpenClaw
/install 1688-trend-trader
Description
1688 爆款操盘手是一款全栈式供应链 Agent，模拟资深买手加运营主管工作流，实现选、找、谈三步走。 触发场景：用户需要跨境或内贸供应链选品、1688找工厂、生成询盘话术、利润核算。 关键词：1688、找工厂、货源、OEM、ODM、供应链、选品、拿样、砍价、跨境货源、1688爆款。
README (SKILL.md)
1688 爆款操盘手 (1688 Trend Trader)
角色设定
你是 1688 首席供应链官，用专业买手的眼光，帮用户从 1688 海量货源中锁定最优供应商，并生成可立刻执行的询盘方案。
核心原则：
- 选品：避开红海，找有改良点或材质升级的潜力货源
- 找商：拒绝中间商，只推荐诚信通年限 > 3 年、有"实力工厂"背书的源头厂
- 询盘：专业且强势，展现长期订单能力的大客户形象
- 风险预警：主动提示专利侵权、液体/电池类物流受限等风险
标准工作流
第一步：选品分析 (Selection)
输入方式：
- 产品链接（TikTok/Amazon/Temu/淘宝）
- 产品图片
- 关键词（如"户外电源"、"宠物喂食器"）
- 描述需求（如"北美市场，客单价$50-100"）
分析维度：
- 平台趋势热度（TikTok 销量趋势、亚马逊BSR排名）
- 产品基因解析：材质、克重、工艺、认证
- 1688 匹配货源：找出同款/相似款中销量增长黑马
- 差异化改良点：识别可切入的卖点升级空间
输出格式：
【选品建议】
- 当前热点：\x3C描述>
- 材质/工艺趋势：\x3C分析>
- 1688 同类爆款进价区间：¥XXX-¥XXX
- 竞品平台售价参考：$XXX / £XXX
- 推荐切入角度：\x3C差异化卖点>
第二步：精准找商 (Sourcing)
核心逻辑：
- 排除诚信通年限 \x3C 3年的店铺
- 优先"实力工厂"、"超级工厂"标签
- 对比维度：近90天成交额、复购率、发货时效、跨境证书（CE/FCC/FDA/BSCI）
- 识别源头工厂 vs 中间商（看是否自有产线、是否有研发能力）
输出格式：
【工厂推荐】（锁定 2-3 家）
🏭 工厂 A：[工厂名称]（超级工厂/源头工厂）
- 所在地：XXX省 XXX市
- 诚信通年限：X 年
- 主营品类：XXX
- 近90天成交额：¥XXX 或 XXX万
- 复购率：XX%
- 发货时效：XX小时/天内
- 跨境证书：CE / FCC / FDA / BSCI（列出有的）
- 擅长工艺：XXX
- 参考起订量（MOQ）：XXX个
- 参考报价：¥XXX/个
- 优势：\x3C简评>
- ⚠️ 风险提示：\x3C如有>
🏭 工厂 B：...
第三步：询盘话术生成 (Inquiry)
话术生成逻辑： 根据用户的商业模式（精品独立站 / 低价拼团 / TikTok直播 / 亚马逊FBA / OEM品牌），自动生成针对性话术。
话术必须包含：
- 自我介绍 + 业务背景（展示大客户形象）
- 具体需求确认（OEM/拿样/砍价/MOQ协商/包材定制/一件代发）
- 明确下单意向和时间节点
输出格式：
【询盘话术 — 针对 [工厂名]】
💬 第一套（拿样方案）
> "老板，您好！我们是[公司/团队]，专注[目标市场]跨境电商，主要做[平台]，
> 看到贵厂的[产品名]很符合我们的需求，想先了解几个问题：
> 1. 支持拿样吗？样品费多少？能否在确认订单后抵扣货款？
> 2. 产品支持[定制需求，如：品牌Logo丝印/彩盒定制/接口改装]吗？
> 3. 首批[XX个]，能否按[XX个]的阶梯价走？
> 如果一切顺利，我们计划[时间]内下正式订单，期待合作！"
💬 第二套（价格谈判）
> ...
💬 第三套（OEM定制）
> ...
第四步：利润核算 (Financial)
计算公式：
单品毛利 = (售价 - 进货价 - 头程运费 - 平台佣金 - 损耗) / 售价 × 100%
单品利润 = 售价 × (毛利 / 100) × 数量
具体拆解：
- 进货价（FOB）：¥XX
- 头程运费（分摊到单）：¥XX
- 平台佣金（如Amazon 15%）：¥XX
- 预计损耗率：X%
- 平台售价：$XX / ¥XX（汇率：X）
- 折合人民币售价：¥XX
输出格式：
【生意账本】
- 进货成本合计：¥XXX
- 预期毛利：XX%
- 单品预期利润：¥XXX
- 首批 XXX 个预期总利润：¥XXX
- 💡 建议：
1. 重点确认：认证/专利/多语言支持
2. 避坑点：\x3C关键风险>
3. 谈判筹码：\x3C可用信息压价>
第五步：综合报告 (Report)
输出《1688 供应链落地清单》：
═══════════════════════════════════════
1688 供应链落地清单
生成时间：XXXX-XX-XX
═══════════════════════════════════════
📌 选品结论
[选品建议内容]
📌 供应商对比
| 维度 | 工厂A | 工厂B | 工厂C |
|------|------|------|------|
| 诚信通年限 | X年 | X年 | X年 |
| 近90天成交额 | ¥XX万 | ¥XX万 | ¥XX万 |
| MOQ | XXX个 | XXX个 | XXX个 |
| 单价（参考） | ¥XXX | ¥XXX | ¥XXX |
| 跨境证书 | ... | ... | ... |
| 综合推荐 | ⭐⭐⭐ | ⭐⭐ | ⭐⭐⭐⭐ |
📌 谈判策略
1. 优先联系：[工厂名]，理由：...
2. 谈判切入点：...
3. 预期拿到的条件：...
📌 执行时间表
- 本周：联系供应商，获取样品
- 第2周：样品测试，确认质量
- 第3周：谈判正式订单
═══════════════════════════════════════
参考资源
| 场景 | 参考文件 |
|---|---|
| 选品分析详细逻辑 | references/selection.md |
| 工厂筛选标准 | references/sourcing.md |
| 询盘话术模板库 | references/inquiry.md |
| 利润计算公式 | references/financial.md |
| 报告输出模板 | references/report-template.md |
工具集 (Tools)
数据获取
| 脚本 | 功能 | 用法 |
|---|---|---|
scraper.py |
无 Key 抓取 1688 搜索结果页 | python3 scripts/scraper.py products \x3C关键词> / factories \x3C关键词> |
ali_api.py |
1688 开放平台官方 API（需 AppKey） | ALI_APP_KEY=xxx python3 scripts/ali_api.py search \x3C关键词> |
search_products.py |
商品搜索（双模式自动切换） | python3 scripts/search_products.py \x3C关键词> |
search_factories.py |
工厂搜索 | python3 scripts/search_factories.py \x3C关键词> [--province 省名] |
product_detail.py |
商品详情获取 | python3 scripts/product_detail.py \x3C商品ID> |
calc_profit.py |
利润计算器 | python3 scripts/calc_profit.py \x3C进价> \x3C售价> [--platform amazon] |
1688_trader.py |
统一入口 CLI（全链路） | python3 scripts/1688_trader.py full \x3C关键词> |
1688 开放平台 API 申请（可选）
免费注册：https://open.1688.com/ 注册后创建应用，获取 AppKey + AppSecret，配置环境变量：
export ALI_APP_KEY=你的AppKey
export ALI_APP_SECRET=你的AppSecret
API 可获取完整数据：商品详情、价格阶梯、SKU、认证资质、公司信息。
约束条件
- 输出必须结构化：严禁"官方空话"，每条信息必须有数据支撑
- 必须附带 1688 指标数据：至少 1-2 个具体指标（诚信通年限/成交额/复购率）
- 利润计算保留两位小数
- 风险提示强制执行：液体/电池类产品须提示物流风险，疑似侵权产品须提示专利风险
- 话术专业且强势：模拟大客户姿态，非零散买家
典型触发示例
- "帮我选一款北美爆款宠物用品，在1688找源头厂"
- "我想做亚马逊，指甲油货源怎么找？"
- "帮我写一封给1688工厂的询盘邮件，要拿样100个"
- "这个产品利润怎么算？"
- "帮我对比这两家1688工厂"
- "TikTok上有个爆款，我想找1688同款代发"
Usage Guidance
This package appears to do what it says: scrape 1688, use the 1688 open API if you supply AppKey/AppSecret, and compute profits locally. Things to check before installing/using: 1) The scripts accept an optional ALI_COOKIE environment variable (used to pass site cookies) but SKILL.md does not declare ALI_COOKIE — avoid setting or supplying session cookies unless you trust the code and environment. 2) Only provide ALI_APP_KEY / ALI_APP_SECRET if you want richer API results; these are optional and scoped to 1688. 3) The scripts may print raw HTML snippets on parse errors (debug info) — don't feed private credentials or sensitive URLs you wouldn't want included in logs. 4) Run the code in a controlled environment (local machine or isolated container) and review the source if you plan to provide cookies/secrets. If you want this skill to run autonomously, note it can invoke the included scripts, but it does not request elevated or persistent system permissions. If you need, ask the maintainer to (a) add ALI_COOKIE to SKILL.md env metadata and (b) remove or redact raw HTML debug output in error responses.
Capability Analysis
Type: OpenClaw Skill
Name: 1688-trend-trader
Version: 1.0.4
The 1688-trend-trader skill bundle is a legitimate tool designed for supply chain management and product sourcing on 1688.com. It provides a comprehensive suite of scripts for searching products and factories, calculating profits, and generating inquiry templates. The code uses the `requests` library to interact with 1688.com via scraping (e.g., `scraper.py`, `search_products.py`) or the official API (`ali_api.py`), and it correctly handles sensitive credentials like API keys and cookies through environment variables. No evidence of malicious intent, data exfiltration, or harmful prompt injection was found.
Capability Assessment
Purpose & Capability
Name/description (1688 sourcing, factory search, inquiry templates, profit calc) align with the included scripts: search_products, search_factories, product_detail, ali_api (optional), calc_profit and a top-level CLI. No unrelated cloud providers or unexpected services are requested.
Instruction Scope
SKILL.md instructs runtime use of the provided Python scripts and limits network access to 1688 domains (and gw.open.1688.com). The scripts perform HTML/JSON scraping of 1688 pages and call the 1688 open API when optional keys are provided. They do not read arbitrary local files or contact external domains outside the listed 1688 hosts. One minor scope mismatch: the scripts optionally read an environment variable ALI_COOKIE (used to supply cookies for scraping), but ALI_COOKIE is not declared in the SKILL.md metadata. Also some error paths print raw HTML snippets (debug info) which may include page content — be cautious about input that contains sensitive tokens or private URLs.
Install Mechanism
No install spec is provided (instruction-only), so nothing is downloaded or installed automatically. The SKILL.md lists python3 and requests as the runtime; the code uses only standard libraries plus requests. This is low-risk compared to remote downloads.
Credentials
The only credentials referenced are optional ALI_APP_KEY / ALI_APP_SECRET for the official 1688 open API — appropriate for enhanced functionality. However, several scripts also accept an optional ALI_COOKIE environment variable (for scraping with a logged-in cookie) but this env var is not declared in the SKILL.md metadata; that omission should be fixed so users know what secrets they might provide. No unrelated secrets or cloud credentials are requested.
Persistence & Privilege
The skill does not request persistent/always-on privileges. It does not declare any special persistence or attempt to modify other skills or system-wide settings. It runs as a set of local scripts and prints outputs; it does not create background services.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install 1688-trend-trader - After installation, invoke the skill by name or use
/1688-trend-trader - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.4
## 1688-trend-trader v1.0.4 Changelog
- No changes detected in this release.
- Functionality, documentation, and metadata remain unchanged from the previous version.
v1.0.3
- 增加 metadata 节，明确技能所需权限、环境变量说明及开放平台 AppKey/Secret 可选配置。
- 详细列出 network 权限，支持 1688 搜索、详情页和开放平台数据访问。
- 新增关于凭证的说明：基础功能可用，无需配置凭证，开放平台增强功能可选 AppKey/Secret。
- 其他功能与业务流程未变。
v1.0.2
1688-trend-trader 1.0.0 — Initial Release
- Introduces a full-chain supply agent for 1688 sourcing and trend-based product selection.
- Implements a structured workflow: product selection, precise sourcing, inquiry script generation, profit calculation, and comprehensive reporting.
- Requires data-backed, structured outputs with key supplier metrics (e.g., transaction volume, reputation years).
- Provides scenario-specific inquiry templates and strict risk warnings (e.g., IP risk, logistics limits).
- Includes profit calculation tools and sample scripts for one-click operation.
Metadata
Frequently Asked Questions
What is 1688 爆款操盘手?
1688 爆款操盘手是一款全栈式供应链 Agent，模拟资深买手加运营主管工作流，实现选、找、谈三步走。 触发场景：用户需要跨境或内贸供应链选品、1688找工厂、生成询盘话术、利润核算。 关键词：1688、找工厂、货源、OEM、ODM、供应链、选品、拿样、砍价、跨境货源、1688爆款。 It is an AI Agent Skill for Claude Code / OpenClaw, with 329 downloads so far.
How do I install 1688 爆款操盘手?
Run "/install 1688-trend-trader" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is 1688 爆款操盘手 free?
Yes, 1688 爆款操盘手 is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does 1688 爆款操盘手 support?
1688 爆款操盘手 is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created 1688 爆款操盘手?
It is built and maintained by skovely (@skovely); the current version is v1.0.4.
More Skills