← Back to Skills Marketplace
bettermen

外卖评价智能监控

by bettermen · GitHub ↗ · v1.0.0 · MIT-0
cross-platform ⚠ suspicious
44
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install food-review-monitor
Description
餐饮外卖评价智能监控。每日自动检查美团外卖、饿了么/淘宝闪购、京东外卖的评价数据,自动识别口味变化、评价变差、配送超时等异常,生成可视化HTML报告并推送提醒。支持CSV导入和API对接两种模式。触发词:外卖评价监控, 评价分析, 差评预警, 口味监测, 配送异常, review monitor, 外卖差评, 顾...
README (SKILL.md)

外卖评价智能监控 🍔📊

概述

面向餐饮行业的全平台外卖评价监控系统。支持美团外卖、淘宝闪购(饿了么)、京东外卖三大平台,自动检测评价数据异常并推送告警。

核心能力

  1. 数据采集 — CSV/Excel导入(通用)、API对接(进阶)
  2. 智能分析 — 情感分析、趋势对比、关键词提取、异常检测
  3. 可视化报告 — 交互式HTML报告,含评分趋势图、差评分布、关键词云
  4. 异常告警 — 自动检测口味变差、评价下滑、配送超时、服务问题

工作流程

数据导入 → 数据清洗 → 多维度分析 → 异常检测 → HTML报告 → 告警推送

触发方式

用户说出以下关键词即可触发:

  • 「检查外卖评价」「外卖评价监控」「分析评价数据」
  • 「看看最近差评」「口味有没有问题」「有没有配送异常」
  • 「生成评价报告」

操作模式

模式1:CSV导入分析(推荐,零门槛)

用户提供CSV/Excel文件路径 → 自动清洗分析 → 生成报告

支持的CSV格式:从美团开店宝、饿了么商家后台、京东商家后台导出的评价数据。

模式2:API对接(需配置)

首次使用:帮助配置API凭证 → 后续自动拉取数据 → 分析报告

模式3:手动粘贴

用户直接粘贴评价文本 → 单条分析

分析维度

维度 检测内容 告警阈值
口味评分 口味相关关键词频率和情感 负面占比>30%告警
综合评分 评分均值趋势 连续3天下降或低于4.0
配送时效 配送超时相关评价 超时提及率>15%告警
服务质量 服务态度、包装等 负面占比>20%告警
差评突增 差评数量同比变化 日差评>历史均值2倍

配置管理

技能首次运行时自动创建配置文件 ~/.food_review_monitor/config.json,支持配置:

  • 店铺名称和ID
  • 各平台API凭证(可选)
  • 告警阈值自定义
  • 历史数据存储路径

使用示例

用户:帮我看看最近外卖评价怎么样
→ 检查是否有历史数据 → 对比分析 → 生成趋势报告

用户:美团外卖最近有没有差评变多
→ 筛选美团平台数据 → 聚焦差评趋势 → 异常检测

用户:设置每天9点自动检查评价
→ 创建定时任务 → 每日自动拉取/分析 → 有问题推送告警

依赖安装

依赖已在技能元数据中声明,首次运行自动安装。或手动安装:

pip install pandas openpyxl jieba snownlp

脚本路径

所有脚本位于技能目录下,使用 {baseDir} 引用:

  • 主入口:{baseDir}/scripts/monitor.py
  • 数据加载:{baseDir}/scripts/data_loader.py
  • 分析引擎:{baseDir}/scripts/analyzer.py
  • 报告生成:{baseDir}/scripts/reporter.py

注意事项

  • CSV导入模式无需任何平台授权,适合所有商家
  • API模式需要先在对应开放平台注册并获取凭证
  • 历史评价数据存储在本地,不会上传到任何服务器
  • 建议至少保留30天历史数据以建立评分基线
Usage Guidance
Install only if you are comfortable with local storage of review data and reports. Avoid entering real platform API keys unless you can protect or manually permission the config file, and treat ~/.food_review_monitor/config.json, data, and reports as sensitive business data.
Capability Tags
requires-sensitive-credentials
Capability Assessment
Purpose & Capability
The CSV/Excel analysis, sentiment checks, anomaly detection, and HTML report generation fit the stated food-delivery review monitoring purpose. API and alerting claims are described, but the inspected code mostly implements local analysis rather than real platform API fetching or push alerts.
Instruction Scope
The trigger phrases are broad, but the actual CLI actions require user-provided files or explicit setup/check commands; no evidence of prompt-injection instructions or hidden role changes was found.
Install Mechanism
The skill declares Python and common data-analysis packages, plus a custom PIP_INDEX_URL mirror. This is disclosed and coherent with the analytics purpose, though users should be aware dependencies may be installed from that configured package index.
Credentials
Local filesystem use is expected for imported review data and generated reports. The skill also stores customer review history under ~/.food_review_monitor/data, which is purpose-aligned but may contain private business/customer data.
Persistence & Privilege
Setup collects optional Meituan/Eleme/JD API keys and saves them in ~/.food_review_monitor/config.json as plaintext JSON, with no chmod, OS keychain use, or clear warning. The documentation also mentions daily scheduled checks, but no scheduler implementation was found in the artifact.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install food-review-monitor
  3. After installation, invoke the skill by name or use /food-review-monitor
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
首个版本:支持美团/饿了么/京东外卖三平台CSV导入,5维异常检测(口味/配送/服务/评分/差评突增),餐饮领域情感词典,ECharts交互式HTML报告
Metadata
Slug food-review-monitor
Version 1.0.0
License MIT-0
All-time Installs 0
Active Installs 0
Total Versions 1
Frequently Asked Questions

What is 外卖评价智能监控?

餐饮外卖评价智能监控。每日自动检查美团外卖、饿了么/淘宝闪购、京东外卖的评价数据,自动识别口味变化、评价变差、配送超时等异常,生成可视化HTML报告并推送提醒。支持CSV导入和API对接两种模式。触发词:外卖评价监控, 评价分析, 差评预警, 口味监测, 配送异常, review monitor, 外卖差评, 顾... It is an AI Agent Skill for Claude Code / OpenClaw, with 44 downloads so far.

How do I install 外卖评价智能监控?

Run "/install food-review-monitor" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is 外卖评价智能监控 free?

Yes, 外卖评价智能监控 is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does 外卖评价智能监控 support?

外卖评价智能监控 is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created 外卖评价智能监控?

It is built and maintained by bettermen (@bettermen); the current version is v1.0.0.

💬 Comments