电商价格比较

从京东、淘宝、天猫、拼多多等中国电商平台抓取商品价格并进行比较分析。当用户需要比较不同电商平台的商品价格、寻找最佳性价比、或监控商品价格变化时使用。支持关键词搜索、商品链接分析、价格历史追踪和购买建议。

Security Scan

The skill claims multi-platform price-comparison and many helper scripts, but the package only includes a single jd_scraper and several documentation files that reference missing scripts; this mismatch (plus instructions to use heavy tools and anti‑scraping tactics) warrants caution before installing or running.

Purpose & Capability

The skill description and SKILL.md claim support for 京东、淘宝、天猫、拼多多 and list multiple helper scripts (price_history.py, batch_compare.py, taobao_scraper.py, tmall_scraper.py, pdd_scraper.py). The provided code manifest contains only scripts/jd_scraper.py (and documentation/reference files). That means the published bundle does not actually include most platform scrapers or the advanced scripts the docs advertise, which is an important inconsistency between claimed purpose and delivered capabilities.

Instruction Scope

SKILL.md and references describe scraping strategies including browser automation (Playwright), API calls, proxies, user-agent rotation, caching, and potential captcha/anti‑scrape handling. Those instructions are consistent with a web-scraping price-comparison tool, but they also explicitly recommend evasion techniques (proxy rotation, UA changes) which increase operational and legal risk. The instructions reference several scripts that are not present, so runtime behavior described in docs cannot be fully verified from included files.

Install Mechanism

There is no install spec (instruction-only), which minimizes automatic install risk. However README and SKILL.md recommend installing Playwright and other Python packages (playwright, beautifulsoup4, requests, etc.). Playwright requires downloading browser binaries (playwright install chromium). The lack of an automated install means the user will perform these steps manually; this is not malicious but is heavier than the manifest implies.

Credentials

The skill declares no required environment variables or credentials, and the included jd_scraper.py uses only public JD endpoints and does not request secrets. The README shows an optional config.yaml (proxy credentials, request settings) — optional proxies or credentialed services could be configured by the user, but they are not required by the skill as published.

Persistence & Privilege

The skill does not request always:true or any elevated persistence. It does not attempt to modify other skills or system configs. Autonomous invocation is allowed by default but is not combined with broad privileges here.

What to consider before installing

Things to consider before installing or running this skill: - Capability mismatch: The docs promise multi‑platform support and many helper scripts, but the package only includes jd_scraper.py. If you need Taobao/Tmall/PDD support, request the missing scripts or source before trusting results. - Dependency burden: The README asks you to install Playwright (and browsers), Requests, BeautifulSoup, etc. Playwright downloads browser binaries—be prepared for that and run installs in a controlled environment. - Anti‑scraping advice: The skill recommends proxies, UA rotation, and other evasion techniques. Those are functionally related to scraping but raise ethical, legal, and operational concerns; avoid using third‑party proxy services you don't trust and ensure scraping complies with each platform's terms of service and local law. - Data & secrets: The skill does not request credentials, but config examples include proxy endpoints and potential login handling for platform-specific member prices. Do not supply platform account credentials to untrusted code. - Test safely: Run the included jd_scraper.py in a sandbox/VM or review the code locally. If you expect full multi‑platform functionality, ask the publisher for the missing script files or a complete source repo and re-evaluate. - If unsure: treat this package as incomplete rather than fully functional; request source provenance or a homepage, and prefer a vetted tool or official APIs for production use.

电商价格比较技能

本技能帮助用户在中国主流电商平台（京东、淘宝、天猫、拼多多）之间进行商品价格比较，找出最佳购买方案。

快速开始

基本使用方式

1. 提供商品信息：可以给商品链接或商品名称
2. 执行价格比较：技能会自动搜索各平台的价格
3. 查看分析结果：获得价格对比和购买建议

示例命令

# 通过商品名称搜索
比较"iPhone 15"在各平台的价格

# 通过商品链接分析
分析这个京东链接的价格：https://item.jd.com/123456.html

# 多商品比较
比较"小米手机"和"华为手机"的价格差异

核心功能

1. 价格抓取

• 京东：支持自营和第三方商家
• 淘宝：支持普通店铺和天猫店铺
• 天猫：支持品牌官方店
• 拼多多：支持百亿补贴和普通商品

2. 数据对比维度

• 价格：商品售价、优惠价、券后价
• 运费：包邮条件、运费金额
• 促销：满减、折扣、赠品
• 评价：评分、评价数量
• 店铺：店铺评分、信誉等级

3. 分析算法

• 价格排序：从低到高排列
• 性价比计算：结合价格、评价、店铺信誉
• 风险评估：识别价格异常或可疑商品
• 推荐方案：给出最佳购买建议

技术实现

抓取策略

由于电商平台的防爬机制，采用混合策略：

1. 浏览器自动化：使用Playwright进行动态页面渲染
2. API调用：使用平台官方API（如可用）
3. 数据缓存：减少重复请求，提高效率

脚本说明

详见 scripts/README.md 了解各平台抓取脚本的详细说明。

使用场景

场景1：日常购物比较

用户想购买某个商品，需要知道哪个平台最便宜。

场景2：价格监控

用户想监控某个商品的价格变化，等待最佳购买时机。

场景3：批量采购

用户需要批量采购商品，需要比较不同供应商的价格。

场景4：市场调研

用户需要了解某个品类在各平台的价格分布。

注意事项

1. 价格实时性：电商价格变化频繁，结果仅供参考
2. 库存状态：低价商品可能缺货或限购
3. 地区差异：价格可能因用户地区不同而变化
4. 账号差异：不同账号可能看到不同价格（如会员价）

高级功能

价格历史追踪

使用 scripts/price_history.py 可以追踪商品价格变化趋势。

批量比较

使用 scripts/batch_compare.py 可以同时比较多个商品。

自定义权重

在 references/comparison_logic.md 中可以自定义价格比较的权重设置。

故障排除

常见问题

1. 抓取失败：平台防爬机制更新，需要更新脚本
2. 价格不一致：不同时间、不同账号看到的价格可能不同
3. 商品匹配错误：不同平台的商品规格可能不完全相同

解决方案

1. 检查网络连接和代理设置
2. 尝试使用不同的用户代理
3. 手动验证商品链接是否有效

相关资源

• 平台API文档 - 各电商平台的API信息
• 比较逻辑说明 - 价格比较算法详解
• 报告模板 - 分析报告模板