An intelligent, AI-powered system that automates the entire LinkedIn content creation workflow—from research to publication—with human approval gates using Trello.

• 🔍 AI-Powered Research: Automatically discovers trending topics based on your target URL and industry
• 📝 Content Generation: Uses CrewAI agents to create engaging, professional LinkedIn posts
• ✅ Approval Workflow: Integration with Trello for manual review and approval at each stage
• 🤖 Automated Scheduling: Daily research and periodic checking for approvals
• 📊 Smart Orchestration: Seamlessly manages the entire workflow from research to publication
• 🎯 Industry-Specific: Tailored content for your specific industry and audience

• Python 3.9 or higher
• OpenAI API key
• Trello account with API access
• LinkedIn Developer App

1. Clone the repository
2. Setup venv: python -m venv venv && source venv/bin/activate
3. Install deps: pip install -r requirements.txt
4. Setup config: cp .env.example .env and edit your keys.

This project supports automated token refreshing. To set it up:

1. Create a LinkedIn App and enable 'Share on LinkedIn' and 'Sign In with LinkedIn'.
2. Add http://localhost:8000/callback to your authorized redirect URLs.
3. Run the helper to generate your tokens:

   python get_linkedin_token.py

4. Copy the ACCESS_TOKEN and REFRESH_TOKEN into your .env file.

The system will now automatically refresh your access token whenever it expires.

Auto LinkedIN/
├── agents/             # AI CrewAI Agents
├── integrations/       # API Clients (LinkedIn, Trello)
├── utils/              # Helper utilities
├── orchestrator.py     # Main Workflow Logic
├── main.py             # CLI Entry Point
└── get_linkedin_token.py # Auth Helper

• Validate: python main.py validate
• Full Run: python main.py full
• Scheduler: python main.py schedule

Built with ❤️ for LinkedIn Automation