1. Project Overview

Lazy Customer Service

Lazy Customer Service is an integrated customer service system built around a knowledge base powered by LLMs (Large Language Models). It provides an out-of-the-box intelligent customer service solution and supports integration with platforms such as WeChat, WeCom, Qianniu, Bilibili, Douyin Enterprise Accounts, Douyin, Douyin Shop, Weibo Chat, Xiaohongshu Professional Accounts, Xiaohongshu, and Zhihu. It supports text, voice, and images, and can access and operate external resources such as operating systems and the internet through plugins. It also supports customizing enterprise AI applications based on your own knowledge base.

2. Open Source License

Licensed under AGPL-3.0

3. Screenshots

4. Feature Overview

Main Functions

• Multi-Platform Support: Currently supports platforms such as WeChat, WeCom, Qianniu, Bilibili, Douyin Enterprise Accounts, Douyin, Douyin Shop, Weibo Chat, Xiaohongshu Professional Accounts, Xiaohongshu, and Zhihu. Support for additional social media platforms will continue to expand in the future.

• Preset Replies: Allows users to configure custom replies to handle frequently asked questions and improve response efficiency.

• ChatGPT Integration: Connects to the ChatGPT API to intelligently generate replies based on customer inquiries. Suitable for handling complex or personalized customer requests.

• Send Images and Binary Files: Supports sending images and other binary files to meet diverse customer service needs.

• Knowledge Base: Create a dedicated chatbot by uploading knowledge base files. It can be used as a digital twin, intelligent customer service agent, or private-domain assistant.

• Independent plugin systems for each platform, supporting plugin access to external resources such as the operating system and the Internet. Also supports customizing enterprise AI applications based on proprietary knowledge bases.

Note: This project is only a customer service tool for connecting to external platforms. It does not provide accounts for any platforms, so you must prepare your own accounts. You must also prepare your own knowledge base files. Currently, this project only supports integration with the OpenAI API (or compatible proxy).

Project Roadmap

• Support basic chat functionality for WeChat.

• Support basic chat functionality for Qianniu.

• Support basic chat functionality for Jingmai.

• Add an option to disable automatic replies and use keyword-based replies only.

• Add randomized reply delays.

• Improve independent configuration options for each platform.

• Support importing reply content from Excel.

• Support exporting reply content to Excel.

• Improve the Qianniu integration mode.

• Improve reply speed on the WeChat platform.

• Add support for Chinese file paths.

• Support external group chats on WeCom.

• Automatically detect human activity and pause the project.

• Add a feature requiring users to @mention the bot in wx group chats to trigger replies.

• Support multiple browser instances.

• Support exporting chat history to Excel.

• Add keyword matching testing functionality.

• Implement automatic private messaging after replying on Xiaohongshu.

• Automatically identify internal group chats on WeCom.

• Support automatically adding contacts on WeCom.

• Support automatic replies on the Douyin Live platform.

• Support Pinduoduo.

• Support automatic article publishing across multiple platforms.

• Support loading local chat models.

5. Tech Stack

This project is an intelligent conversational customer service tool powered by large language models. It supports integration with platforms such as Bilibili, Douyin Enterprise Accounts, Douyin, Douyin Shop, Weibo Chat, Xiaohongshu Professional Accounts, Xiaohongshu, and Zhihu. You can choose GPT3.5/GPT4.0/[Lazy Toolbox] (more platforms will be supported in the future). It can process text, voice, and images, access external resources such as the operating system and the Internet through plugins, and support customized enterprise AI applications based on proprietary knowledge bases.

Set Up the Local Environment

• electron-react-boilerplate

• Node.js v18.x (The latest version is not recommended due to potential compatibility issues.)

• pnpm version 8.x.x

• Git

For more information, please refer to the README.md document