Message Integrate Agent | 消息通信中枢 Agent

English | 中文

English

Overview

Message Integrate Agent is an AI-powered intelligent message hub that connects to multiple messaging platforms (Telegram, Feishu/Lark, WeChat) with intelligent task routing, self-evolution capabilities, and enterprise-grade Docker deployment support.

Inspired by OpenClaw and designed with production-grade architecture, it serves as a unified gateway for multi-platform message management with AI-driven automation.

Key Features

🔌 Multi-Platform Support

Telegram - Bot API with webhook support
Feishu/Lark - WebSocket long connection + webhook (enterprise-ready)
WeChat - Webhook integration
Unified message format across all platforms

🧠 Intelligent Routing

Keyword Router - Fast rule-based routing (<10ms)
AI Intent Router - LLM-powered semantic understanding
Multi-level Pipeline - Menu → Rules → Vector → LLM fallback

🔄 Self-Evolution Engine (Heartbeat)

Inspired by OpenClaw's autonomous agent design:

Information Intake - Tiered information gathering (Hot → RSS → Academic → Search)
Value Judgment - AI-powered content quality scoring
Knowledge Output - Structured learning storage
Social Maintenance - Platform health monitoring
Self-Reflection - Performance analysis and optimization

📡 Intelligence Pipeline

Tier 1: Hot Trends - Hacker News, GitHub Trending, Weibo (direct API, no proxy)
Tier 2: RSS Feeds - 300+ curated sources (WorldMonitor-style)
Tier 3: Academic Papers - arXiv, PubMed API
Tier 4: AI Search - Tavily (as supplement only)

🐳 Docker-Ready

One-click NAS deployment
Built-in proxy support (mihomo/Clash)
Health checks and auto-restart
Resource limits and security hardening

🔧 MCP Integration

Model Context Protocol for tool exposure
Search, LLM, Weather tools readily available

📊 Sentiment Analysis

Sentiment analysis for stored intelligence
Positive/Negative/Neutral classification
LLM-powered deep analysis
Usage: analyze id=123456

🔮 Prediction Analysis

Future trend prediction based on content
Supports URL or text input
Multi-dimensional scenario projection with probability assessment
Usage: predict https://news.example.com or predict AI technology trends

Architecture

┌─────────────────────────────────────────────────────────────────┐
│                        User Layer                                │
│   Telegram    │    Feishu    │    WeChat    │   WebSocket    │
└──────────────────────────┬──────────────────────────────────────┘
                           │
                           ▼
┌─────────────────────────────────────────────────────────────────┐
│                      Gateway Layer                               │
│  ┌─────────────┐  ┌──────────────┐  ┌───────────────────┐   │
│  │   WebSocket │  │   Webhook   │  │  Message Pipeline │   │
│  │   Server    │  │   Handler   │  │  Router → Agent  │   │
│  └─────────────┘  └──────────────┘  └───────────────────┘   │
└──────────────────────────┬──────────────────────────────────────┘
                           │
        ┌─────────────────┼─────────────────┐
        ▼                 ▼                 ▼
┌──────────────┐  ┌──────────────┐  ┌──────────────┐
│  Agent Pool  │  │  Agent Loop  │  │  MCP Server  │
│  (Fast Path)│  │ (Deep Think)│  │  (Tools)    │
└──────────────┘  └──────────────┘  └──────────────┘

┌─────────────────────────────────────────────────────────────────┐
│                   Self-Evolution Layer                          │
│  Heartbeat Engine  │  Skills Loader  │  Self-Learning Router │
└─────────────────────────────────────────────────────────────────┘

Quick Start

1. Clone & Install

git clone https://github.com/mightyoung/message-integrate-agent.git
cd message-integrate-agent

# Create virtual environment
python -m venv .venv
source .venv/bin/activate  # Linux/Mac
# .venv\Scripts\activate   # Windows

# Install dependencies
pip install -r requirements.txt

2. Configure

cp .env.example .env
# Edit .env with your API keys

# Required: OPENAI_API_KEY
# Optional: TELEGRAM_BOT_TOKEN, FEISHU_APP_ID, TAVILY_API_KEY, etc.

3. Run

# Local development
python -m src.main

# Or use startup script
./start.sh

4. Docker Deployment (Recommended for Production)

# Build Docker image
docker build -f Dockerfile.prod -t message-integrate-agent:prod .

# Deploy with docker-compose
cd docker-images
docker-compose -f docker-compose.prod.yml up -d

Configuration

Environment Variables

Variable	Required	Description
`OPENAI_API_KEY`	✅	OpenAI API key for LLM
`FEISHU_APP_ID`	-	Feishu app ID
`FEISHU_APP_SECRET`	-	Feishu app secret
`TELEGRAM_BOT_TOKEN`	-	Telegram bot token
`TAVILY_API_KEY`	-	Tavily search API key
`HTTP_PROXY`	-	HTTP proxy URL
`HTTPS_PROXY`	-	HTTPS proxy URL

Project Structure

message-integrate-agent/
├── src/
│   ├── adapters/          # Platform adapters (Telegram, Feishu, WeChat)
│   ├── agents/            # Agent pool and loop
│   ├── gateway/           # WebSocket gateway and routing
│   ├── heartbeat/         # Self-evolution engine
│   ├── intelligence/      # Intelligence pipeline (RSS, hot, papers)
│   ├── router/           # Intent routing
│   ├── skills/           # Dynamic skill loader
│   └── storage/          # Storage clients (Redis, PostgreSQL, S3)
├── docker-images/         # Docker deployment files
├── docs/                  # Architecture and design docs
├── tests/                 # Test suite
└── scripts/               # Deployment scripts

License

MIT License - see LICENSE for details.

中文

概述

消息通信中枢 Agent (Message Integrate Agent) 是一个 AI 驱动的智能消息中枢，连接多个消息平台（ Telegram、飞书、微信），具备智能任务路由、自我进化能力和企业级 Docker 部署支持。

设计参考 OpenClaw 自主智能体架构，采用生产级架构设计，作为多平台消息管理的统一入口，实现 AI 驱动的自动化。

核心特性

🔌 多平台支持

Telegram - Bot API + Webhook
飞书/钉钉 - WebSocket 长连接 + Webhook（企业级）
微信 - Webhook 集成
统一消息格式

🧠 智能路由

关键词路由 - 快速规则匹配 (<10ms)
AI 意图路由 - LLM 语义理解
多级管道 - 菜单 → 规则 → 向量 → LLM 兜底

🔄 自我进化引擎 (Heartbeat)

参考 OpenClaw 自主智能体设计：

信息摄入 - 分层信息获取 (热榜 → RSS → 学术 → 搜索)
价值判断 - AI 内容质量评分
知识输出 - 结构化学习存储
社交维护 - 平台健康监控
自我反思 - 性能分析与优化

📡 情报流水线

Tier 1: 热榜 - Hacker News、GitHub Trending、微博热搜 (直连 API，无需代理)
Tier 2: RSS 订阅 - 300+ 精选源 (类 WorldMonitor)
Tier 3: 学术论文 - arXiv、PubMed API
Tier 4: AI 搜索 - Tavily (仅补充)

🐳 Docker 部署就绪

一键 NAS 部署
内置代理支持 (mihomo/Clash)
健康检查与自动重启
资源限制与安全加固

🔧 MCP 集成

Model Context Protocol 工具暴露
搜索、LLM、天气等工具开箱即用

📊 舆情分析

对已存储情报进行情感分析
支持正向/负向/中性情感判断
使用 LLM 进行深度分析
使用方式: 分析 id=123456

🔮 预测分析

基于内容进行未来趋势预测
支持 URL 或文本输入
多维度情景推演与概率评估
使用方式: 预测 https://news.example.com 或 预测某项技术的发展趋势

架构

┌─────────────────────────────────────────────────────────────────┐
│                         用户层                                    │
│   Telegram    │    飞书    │    微信    │   WebSocket        │
└──────────────────────────┬──────────────────────────────────────┘
                           │
                           ▼
┌─────────────────────────────────────────────────────────────────┐
│                        网关层                                      │
│  ┌─────────────┐  ┌──────────────┐  ┌───────────────────┐      │
│  │  WebSocket │  │   Webhook   │  │  消息处理管道     │      │
│  │   服务器    │  │   处理器    │  │  路由 → Agent   │      │
│  └─────────────┘  └──────────────┘  └───────────────────┘      │
└──────────────────────────┬──────────────────────────────────────┘
                           │
        ┌─────────────────┼─────────────────┐
        ▼                 ▼                 ▼
┌──────────────┐  ┌──────────────┐  ┌──────────────┐
│  Agent 池   │  │  Agent Loop  │  │  MCP Server  │
│  (快速路径) │  │  (深度思考)  │  │   (工具)     │
└──────────────┘  └──────────────┘  └──────────────┘

┌─────────────────────────────────────────────────────────────────┐
│                       自我进化层                                  │
│  Heartbeat 引擎  │  技能加载器  │  自学习路由                    │
└─────────────────────────────────────────────────────────────────┘

快速开始

1. 克隆与安装

git clone https://github.com/mightyoung/message-integrate-agent.git
cd message-integrate-agent

# 创建虚拟环境
python -m venv .venv
source .venv/bin/activate  # Linux/Mac
# .venv\Scripts\activate   # Windows

# 安装依赖
pip install -r requirements.txt

2. 配置

cp .env.example .env
# 编辑 .env 填写 API 密钥

# 必需: OPENAI_API_KEY
# 可选: TELEGRAM_BOT_TOKEN, FEISHU_APP_ID, TAVILY_API_KEY 等

3. 运行

# 本地开发
python -m src.main

# 或使用启动脚本
./start.sh

4. Docker 部署（生产环境推荐）

# 构建 Docker 镜像
docker build -f Dockerfile.prod -t message-integrate-agent:prod .

# 使用 docker-compose 部署
cd docker-images
docker-compose -f docker-compose.prod.yml up -d

配置说明

环境变量

变量	必需	说明
`OPENAI_API_KEY`	✅	OpenAI API 密钥
`FEISHU_APP_ID`	-	飞书应用 ID
`FEISHU_APP_SECRET`	-	飞书应用密钥
`TELEGRAM_BOT_TOKEN`	-	Telegram Bot 令牌
`TAVILY_API_KEY`	-	Tavily 搜索 API 密钥
`HTTP_PROXY`	-	HTTP 代理地址
`HTTPS_PROXY`	-	HTTPS 代理地址

项目结构

message-integrate-agent/
├── src/
│   ├── adapters/          # 平台适配器 (Telegram, 飞书, 微信)
│   ├── agents/            # Agent 池和循环
│   ├── gateway/          # WebSocket 网关和路由
│   ├── heartbeat/        # 自我进化引擎
│   ├── intelligence/     # 情报流水线 (RSS, 热榜, 论文)
│   ├── router/           # 意图路由
│   ├── skills/           # 动态技能加载器
│   └── storage/          # 存储客户端 (Redis, PostgreSQL, S3)
├── docker-images/         # Docker 部署文件
├── docs/                  # 架构和设计文档
├── tests/                 # 测试套件
└── scripts/               # 部署脚本

License

MIT License - 详见 LICENSE

Star ⭐ | Fork 🍴 | Contribute 🤝

Name		Name	Last commit message	Last commit date
Latest commit History 35 Commits
.claude/skills/gitnexus		.claude/skills/gitnexus
.github/workflows		.github/workflows
.learnings		.learnings
bettafish		bettafish
config		config
docker-images		docker-images
docs		docs
logs		logs
message-hub-20260309		message-hub-20260309
mirofish		mirofish
scripts		scripts
src		src
tests		tests
.dockerignore		.dockerignore
.env.example		.env.example
.env.prod.example		.env.prod.example
.gitignore		.gitignore
AGENTS.md		AGENTS.md
CLAUDE.md		CLAUDE.md
DEPLOY.md		DEPLOY.md
Dockerfile		Dockerfile
Dockerfile.prod		Dockerfile.prod
README-DEPLOY.md		README-DEPLOY.md
README.md		README.md
claude.settings.json		claude.settings.json
deploy.sh		deploy.sh
docker-compose.prod.yml		docker-compose.prod.yml
docker-compose.yml		docker-compose.yml
findings.md		findings.md
message-hub-deploy.tar.gz		message-hub-deploy.tar.gz
package.sh		package.sh
progress.md		progress.md
pyproject.toml		pyproject.toml
requirements.txt		requirements.txt
start-docker.sh		start-docker.sh
start.sh		start.sh
stop.sh		stop.sh
task_plan.md		task_plan.md
test_academic_prompts.py		test_academic_prompts.py
test_feishu_ws.py		test_feishu_ws.py
test_firecrawl_science.py		test_firecrawl_science.py
test_github_trending.py		test_github_trending.py
test_pipeline_academic.py		test_pipeline_academic.py
test_system.py		test_system.py
test_ws_long_connection.py		test_ws_long_connection.py
webhook_server.py		webhook_server.py

Folders and files

Latest commit

History

Repository files navigation

Message Integrate Agent | 消息通信中枢 Agent

English

Overview

Key Features

🔌 Multi-Platform Support

🧠 Intelligent Routing

🔄 Self-Evolution Engine (Heartbeat)

📡 Intelligence Pipeline

🐳 Docker-Ready

🔧 MCP Integration

📊 Sentiment Analysis

🔮 Prediction Analysis

Architecture

Quick Start

1. Clone & Install

2. Configure

3. Run

4. Docker Deployment (Recommended for Production)

Configuration

Environment Variables

Project Structure

License

中文

概述

核心特性

🔌 多平台支持

🧠 智能路由

🔄 自我进化引擎 (Heartbeat)

📡 情报流水线

🐳 Docker 部署就绪

🔧 MCP 集成

📊 舆情分析

🔮 预测分析

架构

快速开始

1. 克隆与安装

2. 配置

3. 运行

4. Docker 部署（生产环境推荐）

配置说明

环境变量

项目结构

License

About

Resources

Uh oh!

Stars

Watchers

Forks

Releases

Packages 0

Uh oh!

Contributors

Uh oh!

Languages

Packages