GhostWriter AI

Let your ghost write your post

An intelligent social media automation system that transforms news articles into engaging social media content using cutting-edge AI services.

🚀 Overview

GhostWriter AI is a comprehensive automation system that:

• Automatically processes news articles from any source
• Generates optimized social media content using local Ollama AI
• Orchestrates multiple AI services for content analysis and generation
• Provides seamless integration with Telegram for team notifications

Let your ghost write your post - while you focus on strategy, GhostWriter handles the content creation.

🏗️ Architecture

Core Components

┌─────────────────┐    ┌─────────────────┐    ┌─────────────────┐
│   Content       │    │   FastAPI       │    │   Telegram      │
│   Monitor       │───▶│   Orchestrator  │───▶│   Bot           │
│   (Google Apps) │    │   (Backend)     │    │   Notifications │
└─────────────────┘    └─────────────────┘    └─────────────────┘
                              │
                    ┌─────────┼─────────┐
                    ▼         ▼         ▼
              ┌──────────┐┌──────────┐┌──────────┐
              │  Ollama  ││ FastVLM  ││ Content  │
              │ LLM      ││ Vision   ││Generator │
              │ Service  ││ Service  ││ Service  │
              └──────────┘└──────────┘└──────────┘

Team Roles

• Lead Developer: FastAPI Backend Orchestrator
• AI Engineer: Ollama/LLM Integration Specialist
• Vision Engineer: FastVLM Vision Processing Service
• DevOps Engineer: Deployment & Integration Management

✨ Features

🤖 AI-Powered Content Generation

• Article Relevance Analysis: Automatically filters relevant news using local Ollama
• Image Processing: AI-powered image ranking and caption generation
• Content Optimization: Smart content generation for different social platforms

🔧 Service Orchestration

• Dynamic Service Switching: Mock/Real AI service toggling
• Robust Error Handling: Graceful fallbacks and timeout management
• Health Monitoring: Real-time service status and integration monitoring

🔐 Enterprise Security

• API Key Authentication: Secure access control
• Environment-based Configuration: Secure credential management
• Rate Limiting: API abuse prevention

📱 Team Integration

• Telegram Notifications: Real-time alerts via custom bot
• Approval Workflow: Human-in-the-loop content review
• Multi-platform Distribution: Ready for Facebook, Instagram, Twitter

🚀 Quick Start

Prerequisites

• Python 3.9+
• Ollama running locally (127.0.0.1:11434)
• Telegram bot credentials

Installation

1. Clone the repository

   git clone https://github.com/Lucian-Adrian/ghostwriter.git
   cd ghostwriter

2. Create virtual environment

   python -m venv venv
   source venv/bin/activate  # On Windows: venv\Scripts\activate

3. Install dependencies

   cd backend_server
   pip install -r requirements.txt

4. Configure environment

   cp .env.example .env
   # Edit .env with your credentials

5. Start the server

   uvicorn app.main:app --host 0.0.0.0 --port 8003 --reload

Environment Variables

# API Security
API_KEY=your_secure_api_key
AGENTS_API_KEY=your_agents_api_key

# Service URLs
FASTVLM_API_URL=http://localhost:8001
GEMMA9B_API_URL=http://127.0.0.1:11434

# Telegram Bot
TELEGRAM_BOT_TOKEN=your_bot_token
TELEGRAM_CHAT_ID=your_chat_id
TELEGRAM_BOT_USERNAME=@YourBot

# Development
DEBUG=True

📚 API Documentation

Core Endpoints

Process Article

POST /api/v1/process-article
Content-Type: application/json
X-API-Key: your_api_key

{
  "article_url": "https://fmf.md/article-url",
  "source": "fmf_monitor"
}

Health Check

GET /health

Service Management

POST /api/v1/config/switch-to-real-services
POST /api/v1/config/switch-to-mock-services
GET /api/v1/config/test-ollama-connection

Full API Reference

• 📖 Complete API Documentation
• 🏗️ Architecture Guide
• 🚀 Deployment Guide

🧪 Testing

Run Integration Tests

cd backend_server
python test_complete_integration.py

Health Check

curl http://localhost:8003/health

📊 System Status

Core Components

• ✅ Backend Orchestrator: FastAPI server operational
• ✅ Telegram Integration: Bot notifications configured
• ✅ Ollama LLM Service: Local AI processing ready
• 🟡 Vision Service: FastVLM integration ready for setup
• 🟡 Content Monitor: Google Apps Script ready for deployment

Service Endpoints

• Backend Orchestrator: http://localhost:8003
• Ollama LLM: http://127.0.0.1:11434
• Artiom's FastVLM: http://localhost:8001 (configurable)
• Telegram Bot: @OfficialGhostWriterBot

🛠️ Development

Project Structure

ghostwriter/
├── README.md                    # Main documentation
├── CONTRIBUTING.md              # Development guidelines
├── .gitignore                   # Git ignore rules
├── backend_server/
│   ├── app/
│   │   ├── main.py              # FastAPI application
│   │   ├── telegram_handler.py  # Telegram integration
│   │   ├── services/            # AI service integrations
│   │   │   ├── language_service.py # Ollama LLM integration
│   │   │   └── vision_service.py   # FastVLM integration
│   │   └── tools/               # Utility functions
│   │       ├── scraper_tool.py  # Web scraping
│   │       └── image_editor_tool.py # Image processing
│   ├── requirements.txt         # Python dependencies
│   ├── .env.example            # Environment template
│   ├── run_simple.py           # Simple server startup
│   ├── start.bat / start.sh    # Platform-specific startup
│   └── processed_images/       # Image cache directory
└── docs/
    ├── API.md                  # API documentation
    ├── ARCHITECTURE.md         # Technical architecture
    ├── DEPLOYMENT.md           # Deployment guide
    ├── examples.md             # Content examples
    ├── google-apps-script.md   # GAS integration
    └── monitor.gs              # Google Apps Script

Adding New Services

1. Create service module in app/services/
2. Implement service interface
3. Add configuration to .env
4. Update orchestrator in main.py
5. Write comprehensive tests

Contributing

1. Fork the repository
2. Create feature branch: git checkout -b feature/amazing-feature
3. Commit changes: git commit -m 'Add amazing feature'
4. Push to branch: git push origin feature/amazing-feature
5. Open Pull Request

🔧 Configuration

Service Switching

Switch between mock and real services:

# Switch to real services
curl -X POST http://localhost:8003/api/v1/config/switch-to-real-services \
  -H "X-API-Key: your_api_key"

# Switch to mock services  
curl -X POST http://localhost:8003/api/v1/config/switch-to-mock-services \
  -H "X-API-Key: your_api_key"

Health Monitoring

# Check all services
curl http://localhost:8003/health

# Test Ollama connection
curl -X GET http://localhost:8003/api/v1/config/test-ollama-connection \
  -H "X-API-Key: your_api_key"

📞 Support

Documentation

• 🐛 Bug Reports: Create GitHub issue with reproduction steps
• 💡 Feature Requests: Open GitHub discussion
• 📖 Documentation: Check the docs/ folder for detailed guides

📄 License

This project is open source software for AI-powered social media automation.

🎯 Roadmap

Phase 1: Core System

• Backend orchestrator
• Telegram bot integration
• Ollama LLM integration
• Vision service integration
• Content monitoring setup

Phase 2: Production Ready

• Google Apps Script integration
• Production server deployment
• Performance monitoring
• Advanced error handling

Phase 3: Advanced Features

• Multi-platform social media posting
• Content analytics dashboard
• A/B testing for optimization
• Automated content scheduling

---

GhostWriter AI - Let your ghost write your post

Built with ❤️ for intelligent content automation