Files
你的用户名 237c7802e5
Some checks failed
Deploy / deploy (push) Has been cancelled
Initial commit: Telegram Management System
Full-stack web application for Telegram management
- Frontend: Vue 3 + Vben Admin
- Backend: NestJS
- Features: User management, group broadcast, statistics

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude <noreply@anthropic.com>
2025-11-04 15:37:50 +08:00

5.1 KiB
Raw Permalink Blame History

Claude Code × Telegram (gramJS) Marketing Intelligence Agent

🚀 Overview

A comprehensive marketing intelligence system that combines Claude AI with Telegram automation for intelligent campaign management. The system provides full-chain automation from goal setting to strategy execution, with human-in-the-loop controls for high-risk decisions.

🏗️ Architecture

Core Components

  1. Orchestrator Service

    • Task scheduling and state management
    • Redis-based task queue
    • State machine workflow engine
    • Task dependency management
  2. Claude Agent

    • Function calling integration
    • Prompt template management
    • Context and conversation history
    • Multi-model support
  3. gramJS Adapter

    • Telegram API integration
    • Account pool management
    • Message automation tools
    • Event handling
  4. Safety Guard

    • Rate limiting and compliance
    • Content filtering
    • Account health monitoring
    • Telegram ToS compliance
  5. Analytics Service

    • Real-time event tracking
    • Funnel analysis
    • User segmentation
    • ROI calculation
  6. A/B Testing Engine

    • Multi-armed bandit
    • Bayesian optimization
    • Dynamic traffic allocation
    • Experiment management

🛠️ Tech Stack

Backend

  • Language: Node.js (ES6+)
  • Framework: Hapi.js
  • Database: PostgreSQL (main), MongoDB (events), Redis (cache/queue)
  • Message Queue: RabbitMQ
  • Search: Elasticsearch

Infrastructure

  • Container: Docker
  • Orchestration: Kubernetes
  • Monitoring: Prometheus + Grafana
  • Tracing: OpenTelemetry
  • API Gateway: Kong/Nginx

AI/ML

  • LLM: Claude API
  • Vector DB: Pinecone/Weaviate
  • ML Framework: TensorFlow.js

📁 Project Structure

marketing-agent/
├── services/
│   ├── orchestrator/       # Task orchestration service
│   ├── claude-agent/       # Claude AI integration
│   ├── gramjs-adapter/     # Telegram automation
│   ├── safety-guard/       # Compliance and safety
│   ├── analytics/          # Analytics engine
│   ├── ab-testing/         # A/B testing service
│   └── api-gateway/        # API gateway
├── shared/
│   ├── models/            # Shared data models
│   ├── utils/             # Common utilities
│   ├── config/            # Configuration
│   └── types/             # TypeScript types
├── infrastructure/
│   ├── docker/            # Docker configs
│   ├── k8s/               # Kubernetes manifests
│   ├── terraform/         # Infrastructure as code
│   └── scripts/           # Deployment scripts
└── docs/
    ├── api/               # API documentation
    ├── architecture/      # Architecture diagrams
    └── guides/            # User guides

🚦 Getting Started

Prerequisites

  • Node.js 18+
  • Docker & Docker Compose
  • PostgreSQL 14+
  • Redis 7+
  • MongoDB 5+

Installation

# Clone repository
git clone <repository-url>
cd marketing-agent

# Install dependencies
npm install

# Setup environment
cp .env.example .env

# Start services
docker-compose up -d

# Run migrations
npm run migrate

# Start development
npm run dev

🔧 Configuration

Environment Variables

# Claude API
CLAUDE_API_KEY=your_api_key
CLAUDE_MODEL=claude-3-opus-20240229

# Database
POSTGRES_URL=postgresql://user:pass@localhost:5432/marketing
MONGODB_URL=mongodb://localhost:27017/events
REDIS_URL=redis://localhost:6379

# Telegram
TELEGRAM_API_ID=your_api_id
TELEGRAM_API_HASH=your_api_hash

# Services
RABBITMQ_URL=amqp://localhost:5672
ELASTICSEARCH_URL=http://localhost:9200

📋 Features

Campaign Management

  • Goal-driven campaign creation
  • Multi-channel orchestration
  • Budget management
  • Performance tracking

AI-Powered Intelligence

  • Strategy generation
  • Content optimization
  • Audience targeting
  • Predictive analytics

Safety & Compliance

  • GDPR/CCPA compliance
  • Rate limiting
  • Content moderation
  • Audit logging

Human-in-the-Loop

  • Approval workflows
  • Risk assessment
  • Manual override
  • Quality control

🧪 Testing

# Unit tests
npm run test

# Integration tests
npm run test:integration

# E2E tests
npm run test:e2e

# Coverage report
npm run test:coverage

📊 Monitoring

  • Metrics: Prometheus endpoints at /metrics
  • Health: Health checks at /health
  • Logs: Centralized logging via ELK stack
  • Tracing: Distributed tracing with Jaeger

🚀 Deployment

Development

npm run dev

Production

# Build containers
npm run build:docker

# Deploy to Kubernetes
npm run deploy:k8s

# Run migrations
npm run migrate:prod

📖 Documentation

🤝 Contributing

Please read CONTRIBUTING.md for details on our code of conduct and the process for submitting pull requests.

📄 License

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