shaxiu/XianyuAutoAgent: Intelligent Xianyu customer service robot reply system

XianyuAutoAgent is an AI-driven customer service robot system tailored for the Xianyu marketplace, providing uninterrupted 7×24 automated support. It leverages a multi-expert decision engine to enable context-aware conversations, intelligent price negotiation, and sentiment analysis, while integrating web search, RAG knowledge base, and operational monitoring modules. Built on a plugin-based architecture, it supports customizable prompt templates and third-party API integration, and is compatible with mainstream LLM frameworks to significantly boost customer service efficiency and user satisfaction.

shaxiu/XianyuAutoAgent: Intelligent Xianyu customer service robot reply system

🌟 Core Features

Intelligent conversation engine

Functional modules Technical Implementation Key Features
Context-aware Session history storage Lightweight dialogue memory management, with the complete dialogue history as LLM context input
Expert Routing LLM prompt+rule routing Intention recognition based on prompt engineering → Dynamic distribution of expert agents, supporting bargaining/technology/customer service multi-scenario switching

Business Function Matrix

Modules Implemented Planning
Core Engine ✅ LLM automatic reply
✅ Context management
🔄 Sentiment analysis enhancement
Bargaining system ✅ Step price reduction strategy 🔄 Market price comparison function
Technical Support ✅ Web search integration 🔄 RAG knowledge base enhancement
Operation and maintenance monitoring ✅ Basic log 🔄 DingTalk integration
🔄 Web management interface

 

🚴 Quick Start

Environmental requirements

  • Python 3.8+
  • NodeJS 18+

Installation Steps

1. 克隆仓库
git clone https://github.com/shaxiu/XianyuAutoAgent.git
cd xianyu-autoagent

2. 安装依赖
pip install -r requirements.txt

3. 配置环境变量
创建一个 `.env` 文件,包含以下内容,也可直接重命名 `.env.example` :

OPENAI_API_KEY=your_api_key_here
COOKIES_STR=your_cookies_here

注意:代码中OPENAI_API_KEY默认使用的是通义千问,如需使用其他API,请自行修改代码中的模型地址和模型名称;
COOKIES_STR自行在闲鱼网页端获取cookies

4. 创建提示词文件prompts/*_prompt.txt
默认提供四个模板,可自行修改

How to use

Run the main program:

python main.py

Custom prompt words

promptsYou can customize the prompt words for each expert by editing the files in the directory:

  • classify_prompt.txt: Intent classification prompt words
  • price_prompt.txt: Price Expert Tips
  • tech_prompt.txt: Technical Expert Tips
  • default_prompt.txt: Default reply prompt

Libre Depot original article,Publisher:Libre Depot,Please indicate the source when reprinting:https://www.libredepot.top/5614.html

Like (0)
Libre DepotLibre Depot
Previous 11 hours ago
Next 8 hours ago

Related articles

Leave a Reply

Your email address will not be published. Required fields are marked *