XMessageChatAgent

An intelligent chat message processing agent built with Spring Boot, Spring AI, Harness, and LangSmith.

English | 中文

Features

Multi-Platform Chat Integration: Connect to WeChat Work, DingTalk, Telegram, WhatsApp, and Teammors
Message Filtering: Filter chat messages by keywords, time range, sender, etc.
AI-Powered Summarization: Use LLMs like OpenAI, DeepSeek, Doubao for message summarization
Message Forwarding: Forward messages to specified users or groups
Email Notifications: Send messages via email
Redis Integration: Store forwarded messages in Redis for caching
LangSmith Integration: Track and trace LLM interactions
Harness Integration: Orchestrate workflows with Harness

Tech Stack

Java: 21
Spring Boot: 3.2.x
Spring AI: 1.0.0-SNAPSHOT
Spring Data JPA: For database operations
Spring Data Redis: For caching
MySQL: Primary database
Redis: Cache and message queue
OpenAI/DeepSeek/Doubao: LLM providers
LangSmith: LLM tracing
Harness: Workflow orchestration

Quick Start

Prerequisites

Java 21 or higher
Maven 3.8+
MySQL 8.0+
Redis 7.0+
API keys for LLM providers (OpenAI, DeepSeek, Doubao)

Installation

Clone the repository

git clone https://github.com/yourusername/xmessage-chat-agent.git
cd xmessage-chat-agent

Configure environment variables

Create a .env file in the root directory:

# Database Configuration
SPRING_DATASOURCE_URL=jdbc:mysql://localhost:3306/xteam_videos?useUnicode=true&characterEncoding=utf8&serverTimezone=Asia/Shanghai
SPRING_DATASOURCE_USERNAME=your_username
SPRING_DATASOURCE_PASSWORD=your_password

# Redis Configuration
SPRING_DATA_REDIS_HOST=localhost
SPRING_DATA_REDIS_PORT=6379
SPRING_DATA_REDIS_PASSWORD=your_redis_password

# LLM Configuration
OPENAI_API_KEY=your_openai_api_key
DEEPSEEK_API_KEY=your_deepseek_api_key
DOUBAO_API_KEY=your_doubao_api_key

# LangSmith Configuration
LANGSMITH_API_KEY=your_langsmith_api_key
LANGSMITH_PROJECT_NAME=xmessage-chat-agent

# Platform Configuration (Optional)
WECHAT_WORK_ENABLED=false
WECHAT_WORK_CORP_ID=your_corp_id
WECHAT_WORK_CORP_SECRET=your_corp_secret

Run the application

mvn spring-boot:run

The application will start on http://localhost:8080

Configuration

Application Configuration (`application.yml`)

server:
  port: 8080

spring:
  datasource:
    url: ${SPRING_DATASOURCE_URL}
    username: ${SPRING_DATASOURCE_USERNAME}
    password: ${SPRING_DATASOURCE_PASSWORD}
  
  data:
    redis:
      host: ${SPRING_DATA_REDIS_HOST}
      port: ${SPRING_DATA_REDIS_PORT}
      password: ${SPRING_DATA_REDIS_PASSWORD}

  ai:
    active-model: ${AI_ACTIVE_MODEL:openai}
    models:
      openai:
        api-key: ${OPENAI_API_KEY}
        base-url: ${OPENAI_BASE_URL:https://api.openai.com}
        model: gpt-4o
      deepseek:
        api-key: ${DEEPSEEK_API_KEY}
        base-url: ${DEEPSEEK_BASE_URL:https://api.deepseek.com/v1}
        model: deepseek-chat
      doubao:
        api-key: ${DOUBAO_API_KEY}
        base-url: ${DOUBAO_BASE_URL:https://api.doubao.com/v1}
        model: doubao-pro

Chat Platform Configuration

chat:
  data-sources:
    wechat-work:
      enabled: ${WECHAT_WORK_ENABLED:false}
      corp-id: ${WECHAT_WORK_CORP_ID}
      corp-secret: ${WECHAT_WORK_CORP_SECRET}
    dingtalk:
      enabled: ${DINGTALK_ENABLED:false}
      app-key: ${DINGTALK_APP_KEY}
      app-secret: ${DINGTALK_APP_SECRET}
    telegram:
      enabled: ${TELEGRAM_ENABLED:false}
      bot-token: ${TELEGRAM_BOT_TOKEN}
    whatsapp:
      enabled: ${WHATSAPP_ENABLED:false}
      api-key: ${WHATSAPP_API_KEY}
      phone-number-id: ${WHATSAPP_PHONE_NUMBER_ID}
    teammors:
      enabled: ${TEAMMORS_ENABLED:false}
      api-url: ${TEAMMORS_API_URL}
      api-key: ${TEAMMORS_API_KEY}

API Documentation

Agent API

Endpoint	Method	Description
`/api/agent/task`	POST	Execute agent task
`/api/agent/tasks`	GET	Get all tasks
`/api/agent/tasks/{id}`	GET	Get task status
`/api/agent/tasks/{id}`	DELETE	Cancel task

Chat API

Endpoint	Method	Description
`/api/chat/messages`	GET	Get chat messages
`/api/chat/messages/filter`	POST	Filter messages
`/api/chat/summary`	POST	Summarize messages

Forward API

Endpoint	Method	Description
`/api/forward/message`	POST	Forward message
`/api/forward/batch`	POST	Batch forward

Email API

Endpoint	Method	Description
`/api/email/send`	POST	Send email
`/api/email/send/text`	POST	Send email from text

Data Source API

Endpoint	Method	Description
`/api/chat/sources/platforms`	GET	Get supported platforms
`/api/chat/sources/fetch`	POST	Fetch messages from platforms
`/api/chat/sources/send`	POST	Send message to platform

Usage Examples

Execute Agent Task

curl -X POST http://localhost:8080/api/agent/task \
  -H "Content-Type: application/json" \
  -d '{
    "userId": "user123",
    "filter": {
      "chatId": "group_chat_001",
      "isGroup": true,
      "keyword": "重要"
    },
    "needForward": true,
    "forwardTargets": ["user456", "user789"],
    "needEmail": true,
    "emailTargets": ["admin@example.com"],
    "reportTargetUserId": "user123"
  }'

Summarize Messages

curl -X POST http://localhost:8080/api/chat/summary \
  -H "Content-Type: application/json" \
  -d '{
    "chatId": "group_chat_001",
    "customPrompt": "请总结以下群聊消息的要点"
  }'

Fetch Messages from Multiple Platforms

curl -X POST http://localhost:8080/api/chat/sources/fetch \
  -H "Content-Type: application/json" \
  -d '{
    "platforms": ["wechat-work", "dingtalk"],
    "groupId": "group_001",
    "startTime": "2024-01-01",
    "endTime": "2024-01-31"
  }'

Send Message to Platform

curl -X POST http://localhost:8080/api/chat/sources/send \
  -H "Content-Type: application/json" \
  -d '{
    "platform": "wechat-work",
    "targetId": "user123",
    "content": "Hello from XMessageChatAgent!"
  }'

Supported Platforms

Platform	Service Name	Status
WeChat Work	`wechat-work`	✅ Available
DingTalk	`dingtalk`	✅ Available
Telegram	`telegram`	✅ Available
WhatsApp	`whatsapp`	✅ Available
Teammors	`teammors`	✅ Available

Adding a New Platform

Create a new class implementing ChatDataSource:

@Service("newplatform")
public class NewPlatformDataSource implements ChatDataSource {
    // Implement interface methods
}

Add configuration to application.yml:

chat:
  data-sources:
    newplatform:
      enabled: true
      api-key: ${NEW_PLATFORM_API_KEY}

Multi-Model Support

The agent supports multiple LLM providers:

Switch Models at Runtime

curl -X POST http://localhost:8080/api/chat/model/switch \
  -H "Content-Type: application/json" \
  -d '{
    "model": "deepseek"
  }'

Available Models

Model	Provider
`gpt-4o`	OpenAI
`deepseek-chat`	DeepSeek
`doubao-pro`	Doubao

Contributing

We welcome contributions! Please follow these steps:

Fork the repository
Create a feature branch (git checkout -b feature/AmazingFeature)
Commit your changes (git commit -m 'Add some AmazingFeature')
Push to the branch (git push origin feature/AmazingFeature)
Open a Pull Request

Development Guidelines

Follow Java 21 best practices
Use Lombok for reducing boilerplate code
Write unit tests for new features
Follow REST API design principles
Maintain code style consistency

License

This project is licensed under the MIT License - see the LICENSE file for details.

Acknowledgments

XMessageChatAgent

一个基于 Spring Boot、Spring AI、Harness 和 LangSmith 构建的智能聊天消息处理代理。

English | 中文

功能特性

多平台聊天集成: 连接企业微信、钉钉、Telegram、WhatsApp 和 Teammors
消息过滤: 按关键词、时间范围、发送者等条件过滤消息
AI 驱动的总结: 使用 OpenAI、DeepSeek、Doubao 等大模型进行消息总结
消息转发: 将消息转发给指定用户或群聊
邮件通知: 通过邮件发送消息
Redis 集成: 将转发消息存储在 Redis 中进行缓存
LangSmith 集成: 跟踪和追踪 LLM 交互
Harness 集成: 使用 Harness 编排工作流

技术栈

Java: 21
Spring Boot: 3.2.x
Spring AI: 1.0.0-SNAPSHOT
Spring Data JPA: 数据库操作
Spring Data Redis: 缓存
MySQL: 主数据库
Redis: 缓存和消息队列
OpenAI/DeepSeek/Doubao: LLM 提供商
LangSmith: LLM 追踪
Harness: 工作流编排

快速开始

前置条件

Java 21 或更高版本
Maven 3.8+
MySQL 8.0+
Redis 7.0+
LLM 提供商的 API 密钥（OpenAI、DeepSeek、Doubao）

安装步骤

克隆仓库

git clone https://github.com/yourusername/xmessage-chat-agent.git
cd xmessage-chat-agent

配置环境变量

在根目录创建 .env 文件：

# 数据库配置
SPRING_DATASOURCE_URL=jdbc:mysql://localhost:3306/xteam_videos?useUnicode=true&characterEncoding=utf8&serverTimezone=Asia/Shanghai
SPRING_DATASOURCE_USERNAME=your_username
SPRING_DATASOURCE_PASSWORD=your_password

# Redis 配置
SPRING_DATA_REDIS_HOST=localhost
SPRING_DATA_REDIS_PORT=6379
SPRING_DATA_REDIS_PASSWORD=your_redis_password

# LLM 配置
OPENAI_API_KEY=your_openai_api_key
DEEPSEEK_API_KEY=your_deepseek_api_key
DOUBAO_API_KEY=your_doubao_api_key

# LangSmith 配置
LANGSMITH_API_KEY=your_langsmith_api_key
LANGSMITH_PROJECT_NAME=xmessage-chat-agent

# 平台配置（可选）
WECHAT_WORK_ENABLED=false
WECHAT_WORK_CORP_ID=your_corp_id
WECHAT_WORK_CORP_SECRET=your_corp_secret

运行应用

mvn spring-boot:run

应用将在 http://localhost:8080 启动

配置说明

应用配置 (`application.yml`)

server:
  port: 8080

spring:
  datasource:
    url: ${SPRING_DATASOURCE_URL}
    username: ${SPRING_DATASOURCE_USERNAME}
    password: ${SPRING_DATASOURCE_PASSWORD}
  
  data:
    redis:
      host: ${SPRING_DATA_REDIS_HOST}
      port: ${SPRING_DATA_REDIS_PORT}
      password: ${SPRING_DATA_REDIS_PASSWORD}

  ai:
    active-model: ${AI_ACTIVE_MODEL:openai}
    models:
      openai:
        api-key: ${OPENAI_API_KEY}
        base-url: ${OPENAI_BASE_URL:https://api.openai.com}
        model: gpt-4o
      deepseek:
        api-key: ${DEEPSEEK_API_KEY}
        base-url: ${DEEPSEEK_BASE_URL:https://api.deepseek.com/v1}
        model: deepseek-chat
      doubao:
        api-key: ${DOUBAO_API_KEY}
        base-url: ${DOUBAO_BASE_URL:https://api.doubao.com/v1}
        model: doubao-pro

聊天平台配置

chat:
  data-sources:
    wechat-work:
      enabled: ${WECHAT_WORK_ENABLED:false}
      corp-id: ${WECHAT_WORK_CORP_ID}
      corp-secret: ${WECHAT_WORK_CORP_SECRET}
    dingtalk:
      enabled: ${DINGTALK_ENABLED:false}
      app-key: ${DINGTALK_APP_KEY}
      app-secret: ${DINGTALK_APP_SECRET}
    telegram:
      enabled: ${TELEGRAM_ENABLED:false}
      bot-token: ${TELEGRAM_BOT_TOKEN}
    whatsapp:
      enabled: ${WHATSAPP_ENABLED:false}
      api-key: ${WHATSAPP_API_KEY}
      phone-number-id: ${WHATSAPP_PHONE_NUMBER_ID}
    teammors:
      enabled: ${TEAMMORS_ENABLED:false}
      api-url: ${TEAMMORS_API_URL}
      api-key: ${TEAMMORS_API_KEY}

API 文档

Agent API

端点	方法	描述
`/api/agent/task`	POST	执行代理任务
`/api/agent/tasks`	GET	获取所有任务
`/api/agent/tasks/{id}`	GET	获取任务状态
`/api/agent/tasks/{id}`	DELETE	取消任务

聊天 API

端点	方法	描述
`/api/chat/messages`	GET	获取聊天消息
`/api/chat/messages/filter`	POST	过滤消息
`/api/chat/summary`	POST	总结消息

转发 API

端点	方法	描述
`/api/forward/message`	POST	转发消息
`/api/forward/batch`	POST	批量转发

邮件 API

端点	方法	描述
`/api/email/send`	POST	发送邮件
`/api/email/send/text`	POST	从文本发送邮件

数据源 API

端点	方法	描述
`/api/chat/sources/platforms`	GET	获取支持的平台
`/api/chat/sources/fetch`	POST	从平台获取消息
`/api/chat/sources/send`	POST	发送消息到平台

使用示例

执行代理任务

curl -X POST http://localhost:8080/api/agent/task \
  -H "Content-Type: application/json" \
  -d '{
    "userId": "user123",
    "filter": {
      "chatId": "group_chat_001",
      "isGroup": true,
      "keyword": "重要"
    },
    "needForward": true,
    "forwardTargets": ["user456", "user789"],
    "needEmail": true,
    "emailTargets": ["admin@example.com"],
    "reportTargetUserId": "user123"
  }'

总结消息

curl -X POST http://localhost:8080/api/chat/summary \
  -H "Content-Type: application/json" \
  -d '{
    "chatId": "group_chat_001",
    "customPrompt": "请总结以下群聊消息的要点"
  }'

从多个平台获取消息

curl -X POST http://localhost:8080/api/chat/sources/fetch \
  -H "Content-Type: application/json" \
  -d '{
    "platforms": ["wechat-work", "dingtalk"],
    "groupId": "group_001",
    "startTime": "2024-01-01",
    "endTime": "2024-01-31"
  }'

发送消息到平台

curl -X POST http://localhost:8080/api/chat/sources/send \
  -H "Content-Type: application/json" \
  -d '{
    "platform": "wechat-work",
    "targetId": "user123",
    "content": "来自 XMessageChatAgent 的消息！"
  }'

支持的平台

平台	服务名称	状态
企业微信	`wechat-work`	✅ 可用
钉钉	`dingtalk`	✅ 可用
Telegram	`telegram`	✅ 可用
WhatsApp	`whatsapp`	✅ 可用
Teammors	`teammors`	✅ 可用

添加新平台

创建实现 ChatDataSource 接口的类：

@Service("newplatform")
public class NewPlatformDataSource implements ChatDataSource {
    // 实现接口方法
}

在 application.yml 中添加配置：

chat:
  data-sources:
    newplatform:
      enabled: true
      api-key: ${NEW_PLATFORM_API_KEY}

多模型支持

代理支持多个 LLM 提供商：

运行时切换模型

curl -X POST http://localhost:8080/api/chat/model/switch \
  -H "Content-Type: application/json" \
  -d '{
    "model": "deepseek"
  }'

可用模型

模型	提供商
`gpt-4o`	OpenAI
`deepseek-chat`	DeepSeek
`doubao-pro`	Doubao

贡献指南

欢迎贡献代码！请遵循以下步骤：

Fork 仓库
创建功能分支 (git checkout -b feature/AmazingFeature)
提交更改 (git commit -m 'Add some AmazingFeature')
推送到分支 (git push origin feature/AmazingFeature)
打开 Pull Request

开发规范

遵循 Java 21 最佳实践
使用 Lombok 减少样板代码
为新功能编写单元测试
遵循 REST API 设计原则
保持代码风格一致

许可证

本项目采用 MIT 许可证 - 详见 LICENSE 文件。

Name		Name	Last commit message	Last commit date
Latest commit History 1 Commit
skills/send-email		skills/send-email
src/main		src/main
LICENSE		LICENSE
README.md		README.md
ad_banner.png		ad_banner.png
pom.xml		pom.xml

Folders and files

Latest commit

History

Repository files navigation

XMessageChatAgent

Table of Contents

Features

Tech Stack

Quick Start

Prerequisites

Installation

Configuration

Application Configuration (application.yml)

Chat Platform Configuration

API Documentation

Agent API

Chat API

Forward API

Email API

Data Source API

Usage Examples

Execute Agent Task

Summarize Messages

Fetch Messages from Multiple Platforms

Send Message to Platform

Supported Platforms

Adding a New Platform

Multi-Model Support

Switch Models at Runtime

Available Models

Contributing

Development Guidelines

License

Acknowledgments

XMessageChatAgent

目录

功能特性

技术栈

快速开始

前置条件

安装步骤

配置说明

应用配置 (application.yml)

聊天平台配置

API 文档

Agent API

聊天 API

转发 API

邮件 API

数据源 API

使用示例

执行代理任务

总结消息

从多个平台获取消息

发送消息到平台

支持的平台

添加新平台

多模型支持

运行时切换模型

可用模型

贡献指南

开发规范

许可证

致谢

About

Topics

Resources

License

Uh oh!

Stars

Watchers

Forks

Releases

Packages 0

Uh oh!

Contributors

Uh oh!

Languages

Application Configuration (`application.yml`)

应用配置 (`application.yml`)

Packages