← 返回 Skills 市场
购物助手
作者
wuhaichao87
· GitHub ↗
· v1.1.0
· MIT-0
411
总下载
2
收藏
2
当前安装
2
版本数
在 OpenClaw 中安装
/install shopping-assistant
功能描述
智能购物助手,提供淘宝、京东、拼多多优惠券查询、全网比价、价格保护、历史价格及降价提醒服务。
安全使用建议
This skill has multiple red flags: the registry entry includes no code or install steps, but the instructions expect local Python scripts and ask you to supply affiliate credentials and enable invisible backend link conversion. Before installing or providing any credentials: 1) Ask the publisher for source code or a verifiable install package and a homepage or repo; 2) Verify where user links are sent and get a privacy policy/endpoint for the backend that does '转链'; 3) Do not place affiliate keys in ~/.openclaw/.env or run unknown Python scripts until you can review them; 4) Prefer skills that declare required env vars in metadata and include an install spec or code; 5) If you still test it, do so in an isolated environment and monitor network traffic to confirm no unexpected exfiltration. The current state is inconsistent and warrants caution.
功能分析
Type: OpenClaw Skill
Name: shopping-assistant
Version: 1.1.0
The skill bundle describes a shopping assistant for Chinese e-commerce platforms (Taobao, JD, Pinduoduo) that provides coupon searching and price comparison. While the documentation in SKILL.md mentions 'silent' affiliate link conversion for commission tracking, this is a standard business model for such tools and does not indicate malicious intent like data exfiltration or unauthorized system access. No executable code was provided for analysis, and the requested environment variables (e.g., ZHETAOKE_APP_KEY, JD_UNION_ID) are consistent with the stated functionality.
能力评估
Purpose & Capability
The name/description (shopping coupons, price-compare, price-protect) align with needing affiliate/partner keys (ZHETAOKE, JD, TAOBAO). However the skill metadata declares no required env vars or install steps while SKILL.md instructs the user to set multiple affiliate credentials and to run local scripts. Also the skill claims a backend will '转链' (rewrite links for commission) but provides no source, endpoint, or owner/homepage — mismatch between claimed purpose and manifest.
Instruction Scope
SKILL.md tells the agent/user to run python scripts at ~/.openclaw/workspace/skills/shopping-assistant/scripts/*.py and to store affiliate keys in ~/.openclaw/.env. Yet the bundle contains no scripts. The instructions also say links are converted by a backend (用户无感知 — user unaware), implying silent transmission of user-submitted links to an external service for tracking — this is a privacy/consent concern because no endpoints or data handling details are provided.
Install Mechanism
There is no install spec and no code files, but runtime commands refer to scripts that must exist on disk. This inconsistency means either the skill is incomplete (missing install/code) or it expects out-of-band installation from an unknown source — both increase risk because executable code location and provenance are unspecified.
Credentials
The SKILL.md requires affiliate credentials (ZHETAOKE_APP_KEY, ZHETAOKE_SID, JD_UNION_ID, TAOBAO_PID) which are plausible for the described functions, but the registry metadata does not declare any required env vars or a primary credential. Asking users to place these sensitive credentials in ~/.openclaw/.env without manifest declaration or privacy/usage explanation is disproportionate and lacks transparency.
Persistence & Privilege
The skill does not request always:true or other elevated persistence, and there is no install spec that writes to system-wide locations. Autonomous invocation is allowed (platform default). The main concern is not persistence but the missing code and undisclosed backend behavior.
如何使用
- 确保已安装 OpenClaw(本地或 Docker 部署)
- 在对话框中输入安装命令:
/install shopping-assistant - 安装完成后,直接呼叫该 Skill 的名称或使用
/shopping-assistant触发 - 根据 Skill 的参数说明提供必要输入,即可获得结构化输出
版本历史
v1.1.0
添加多平台比价功能,优化查券输出格式
v1.0.0
智能购物助手,支持查券、比价、价保,后台自动转链
元数据
常见问题
购物助手 是什么?
智能购物助手,提供淘宝、京东、拼多多优惠券查询、全网比价、价格保护、历史价格及降价提醒服务。 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 411 次。
如何安装 购物助手?
在 OpenClaw 或 Claude Code 对话框中运行命令「/install shopping-assistant」即可一键安装,无需额外配置。
购物助手 是免费的吗?
是的,购物助手 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。
购物助手 支持哪些平台?
购物助手 跨平台运行,可在任意部署了 OpenClaw / Claude Code 的环境中使用(cross-platform)。
谁开发了 购物助手?
由 wuhaichao87(@wuhaichao87)开发并维护,当前版本 v1.1.0。
推荐 Skills