← 返回 Skills 市场
Product Recommender
作者
fangwei-frank
· GitHub ↗
· v1.0.0
· MIT-0
162
总下载
0
收藏
1
当前安装
1
版本数
在 OpenClaw 中安装
/install product-recommender
功能描述
Intelligent product recommendation engine for retail digital employees. Recommends products based on customer needs, budget, recipient, occasion, preferences...
安全使用建议
This skill appears coherent for recommending products: it runs a local Python script against a provided knowledge_base.json and uses deterministic filtering/scoring. Before installing/using: 1) Review the complete scripts/recommend.py file (the submission truncated the file end) to confirm the main() function does not perform unexpected I/O or network calls. 2) Ensure the agent will be given only trusted knowledge_base JSON files (don't point --kb at sensitive local files). 3) If you rely on inventory/live API behavior, check how 'stock_status: live_api' is handled elsewhere — the skill assumes live availability if that flag is present. 4) Confirm how 'feature_request' logging is implemented in your environment (the docs mention logging but the visible code doesn't show where it goes). If you want a higher assurance, ask for the full recommend.py file (complete) and any runtime wrapper the agent uses to invoke it so you can verify there are no hidden network endpoints or credential usage.
能力评估
Purpose & Capability
Name/description (product recommendation) match the included SKILL.md and the recommend.py logic: intent extraction, budget/constraint filtering, scoring, upsell logic and presentation. Required resources (a products[] knowledge base) are consistent with the stated purpose; no unrelated credentials or binaries are requested.
Instruction Scope
SKILL.md stays within the recommender domain: it instructs extracting intent signals, running scripts/recommend.py, and returning 3 curated items. It does mention logging 'feature_request' and session state (e.g., upsell_declined) which are not fully implemented in the visible code — minor scope mismatch but not evidence of malicious behavior. The runtime instructions do require a knowledge_base.json path; ensure that path only points to intended product data (the script will read any file given).
Install Mechanism
No install spec (instruction-only with an included script). That is low-risk: nothing is downloaded or installed automatically by the skill.
Credentials
The skill requests no environment variables, credentials, or config paths. The code likewise does not reference secrets or external tokens in the visible portion.
Persistence & Privilege
always is false and the skill does not request persistent system-wide privileges. The script appears to be a transient CLI utility that reads a KB and returns recommendations; it does not modify other skills or system settings in the visible code.
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 product-recommender - 安装完成后，直接呼叫该 Skill 的名称或使用
/product-recommender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
v1.0.0
first release
元数据
常见问题
Product Recommender 是什么？
Intelligent product recommendation engine for retail digital employees. Recommends products based on customer needs, budget, recipient, occasion, preferences... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 162 次。
如何安装 Product Recommender？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install product-recommender」即可一键安装，无需额外配置。
Product Recommender 是免费的吗？
是的，Product Recommender 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
Product Recommender 支持哪些平台？
Product Recommender 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 Product Recommender？
由 fangwei-frank（@fangwei-frank）开发并维护，当前版本 v1.0.0。
推荐 Skills