BuyWise

v1.5.2

BuyWise is your personal shopping advisor — it helps you decide whether to buy, where to buy, and when to buy, across all major global and Chinese platforms....

1· 134·0 current·0 all-time

Install

OpenClaw Prompt Flow

Install with OpenClaw

Best for remote or guided setup. Copy the exact prompt, then paste it into OpenClaw for jiajiaoy/buywise.

Previewing Install & Setup.
Prompt PreviewInstall & Setup
Install the skill "BuyWise" (jiajiaoy/buywise) from ClawHub.
Skill page: https://clawhub.ai/jiajiaoy/buywise
Keep the work scoped to this skill only.
After install, inspect the skill metadata and help me finish setup.
Use only the metadata you can verify from ClawHub; do not invent missing requirements.
Ask before making any broader environment changes.

Command Line

CLI Commands

Use the direct CLI path if you want to install manually and keep every step visible.

OpenClaw CLI

Canonical install target

openclaw skills install jiajiaoy/buywise

ClawHub CLI

Package manager switcher

npx clawhub@latest install buywise
Security Scan
Capability signals
CryptoCan make purchases
These labels describe what authority the skill may exercise. They are separate from suspicious or malicious moderation verdicts.
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Benign
high confidence
Purpose & Capability
Name/description (multi‑platform price comparison, review analysis, deal checking) match the code and SKILL.md: scripts and instructions explicitly navigate shopping and price-history sites. No unrelated credentials, binaries, or install steps are requested.
Instruction Scope
SKILL.md and scripts direct the agent to use the browser tool and web_search to open many public pages (smzdm, Google Shopping, Amazon, CamelCamelCamel, AliExpress, Temu, Zhihu, Xiaohongshu, Reddit, etc.) and to extract prices/reviews. This stays within the declared purpose, but it implicitly relies on the agent's browser context: if the browser has account cookies or session state, pages may reveal user-specific data. The skill itself does not instruct reading local files or environment variables.
Install Mechanism
No install spec — instruction-first skill. There is no remote download or archive extraction. This is the lowest-risk installation model and is proportionate for an instruction-driven aggregator.
Credentials
The skill requests no environment variables, credentials, or config paths. It recommends (but does not require) the gstack skill to enable a browser tool; that is coherent because browsing is central to its function. It also references a separate CouponClaw skill for coupon stacking—an expected ecosystem integration rather than hidden credential use.
Persistence & Privilege
always is false and the skill does not request system-wide persistence or modify other skills. Autonomous invocation is permitted by default (platform normal) but not combined with other elevated privileges.
Assessment
BuyWise appears coherent and implements what it promises, but before installing consider: - Browser context: the skill expects to use a browser tool to open shopping sites. If your agent's browser has logged-in accounts, pages may expose personal data (order history, saved addresses). Run the skill in an isolated/clean browser session if you want to avoid exposing account-specific info. - gstack dependency: for full browser navigation, install and review the gstack skill (it grants the agent browsing capability). If you don't install gstack, the skill will fall back to web_search with reduced accuracy. - CouponClaw integration: the skill suggests calling another skill (CouponClaw) for coupon stacking; review that skill's permissions before invoking it. - No credentials requested by BuyWise itself, and there are no remote installs in the package, but always review any downstream skill the tool asks you to run. If you plan to use BuyWise for sensitive purchases, test it first with non‑personal queries and verify the browsing tool's session isolation.

Like a lobster shell, security has layers — review code before you run it.

latestvk9768r9p4psy850e9cnmcyg73h85nh1h
134downloads
1stars
8versions
Updated just now
v1.5.2
MIT-0

BuyWise — 全球购物决策助手

比价 · 识别假促销 · 评价提炼 · 替代品推荐 · 买入时机判断

何时使用

  • 用户问"这个值得买吗""这个怎么样"
  • 用户说"帮我比价""最低价在哪""哪里买最便宜"
  • 用户问"双11能便宜多少""618值得等吗""现在买合适吗"
  • 用户说"帮我看看评价""这个评价怎么样""有没有问题"
  • 用户说"有没有平替""有没有更便宜的""有没有更好的"
  • 用户发来商品链接,询问是否值得购买

🌐 语言规则

  • 默认中文;用户英文提问切英文
  • 价格按用户所在市场显示(国内用 ¥,国际用 $)
  • 平台名称保留原文(不翻译 Amazon / JD / Temu 等)

🛒 覆盖平台

市场平台
中国国内京东、淘宝、天猫、拼多多(百亿补贴)、闲鱼(二手)
国际Amazon、eBay、AliExpress、Temu

🔌 数据来源

BuyWise 通过 browser 工具直接导航到真实数据页面获取价格,不依赖 web_search 猜测:

数据类型来源覆盖平台
中国市场价格什么值得买 smzdm.com京东、淘宝、天猫、拼多多
Amazon 历史价格CamelCamelCamelAmazon 全球
Amazon 当前价格amazon.com 搜索页Amazon
eBay 价格ebay.com 搜索页eBay
AliExpress 价格aliexpress.com 搜索页AliExpress
Temu / 闲鱼web_search 补充Temu、闲鱼
评价web_search + 小红书/知乎/Reddit多平台

需要 gstack skill 已安装以启用 browser 导航能力。未安装时降级为 web_search(精度下降)。


📋 功能说明

综合决策(主功能)

用户说"帮我看看 XX"时触发完整分析:

  1. 多平台比价 — browser 导航 smzdm/Amazon/eBay/AliExpress 等获取真实价格,汇总对比表
  2. 促销真实性 — browser 导航 CamelCamelCamel + smzdm 价格走势图,判断是否真实低价
  3. 评价提炼 — 从京东/知乎/小红书/Reddit 等多源提炼 3 优点 / 3 槽点 / 适合人群 / 红旗警告
  4. 替代品推荐 — 搜索同类更高性价比产品,推荐 2-3 个选项
  5. 购买时机 — 结合历史低价、大促节点、当前价格给出 🟢立即买 / 🟡再等等 / 🔴不建议

独立比价

用户只问"XX 多少钱""哪里最便宜"时,仅输出比价表,不做完整分析。

促销核查

用户说"这个双11真的便宜吗""这个折扣是真的吗"时,专项分析价格历史,判断促销真实性。

评价深读

用户说"详细说说评价""口碑怎么样"时,深度搜索多平台评测、论坛讨论、投诉记录。


🔧 脚本说明

# 完整购物决策(主入口)
node scripts/advise.js <商品名> [--lang zh|en]
# 示例:
node scripts/advise.js "戴森吸尘器 V15"
node scripts/advise.js "Sony WH-1000XM5" --lang en

# 仅比价
node scripts/compare.js <商品名> [--lang zh|en]
node scripts/compare.js "AirPods Pro 2"

# 促销真实性核查
node scripts/deal-check.js <商品名> [--price 当前价] [--was 标称原价] [--lang zh|en]
node scripts/deal-check.js "iPhone 16" --price 5999 --was 7999

# 评价提炼
node scripts/review-scan.js <商品名> [--lang zh|en]
node scripts/review-scan.js "小米路由器 BE7000"

⚠️ 注意事项

  1. 价格基于 WebSearch 实时搜索,可能存在轻微延迟,购买前以平台实际页面为准
  2. 历史价格分析依赖公开信息,若平台不公示价格历史则以搜索结果为准
  3. 评价摘要基于公开评测,不代表所有用户体验
  4. 促销节点判断(双11/618等)以中国电商日历为基准;国际促销参考 Black Friday / Prime Day
  5. 替代品推荐不含广告或赞助,仅基于性价比分析

Comments

Loading comments...