An automated AI customer-service agent for the Xianyu second-hand trading platform that uses multi-expert collaboration for tasks including bargaining and technical support. It requires user-supplied platform cookies and a remote Large Language Model API, and is intended strictly for learning purposes rather than commercial production.
Project overview
It applies a multi-expert collaborative architecture to automate platform-specific buyer-seller communication, including tiered bargaining and context-aware dialogue.
Project type
AI Agent · Prompt Engineering
Use cases
Knowledge Q&A
Deployment
Refer to project documentation
License
GPL-3.0
Best for
General users seeking to understand automated prompt-engineering and multi-expert agent collaboration for the Xianyu platform.
Key capabilities
Manages conversation memory by storing full dialogue history as input context for the language model.
Uses intent recognition via prompt engineering to dynamically dispatch queries to specialized expert agents for scenarios including bargaining, tech support, and general service.
Applies a tiered price reduction strategy during buyer-seller negotiations.
Integrates web search to support technical support scenarios.
Limitations and risks
The project may be stopped or deleted at any time due to its sensitive nature.
The project is intended strictly for learning and communication purposes.
Using platform cookies for automation may violate the platform's terms of service and carries a risk of account bans.
Getting started
Setup involves medium difficulty. Users must manually retrieve platform cookies, configure environment variables, and set API keys. The first success path is to clone the repository, install dependencies via pip, configure the .env file with the API key and cookies, and run the application via python main.py.