🚀 LinkedIn Post Generator with CrewAI

An AI-powered multi-agent system that generates optimized LinkedIn posts by analyzing trends, creating content, enhancing engagement, and reviewing for quality — all automated using CrewAI and Google Gemini.

---

📌 Features

• Multi-Agent Workflow using CrewAI:

  • Trend Analyst → Identifies trending topics in the chosen domain.
  • Content Creator → Crafts engaging and well-structured LinkedIn posts.
  • Engagement Optimizer → Suggests improvements for higher reach (hashtags, tone, hooks).
  • Post Reviewer → Ensures grammar, clarity, and professional tone.

• Sequential AI Pipeline — Tasks run in order for maximum quality.

• Customizable Agents & Tasks via YAML configuration.

• LLM Integration with Google Gemini 2.0 Flash.

📂 Project Structure

linkedin_post_generator/
│
├── main.py                # Main CrewAI workflow
├── config/
│   ├── agents.yaml         # Agent configurations
│   └── tasks.yaml          # Task configurations
├── .env                    # Environment variables
└── requirements.txt        # Python dependencies

Setup

1. Clone the repository:

   git clone https://github.com/Asish-Bothsa/linkedin_post_generator
   cd linkedin_post_generator

2. Install dependencies:
   • Make sure you have Python 3.10+ and uv installed.
   • Install dependencies:

     uv pip install -r requirements.txt
     # or if using pyproject.toml
     uv pip install

3. Set up API keys:
   • Create a .env file in the project root with your LLM API keys:

     OPENAI_API_KEY=your_openai_key_here
     # or
     GOOGLE_API_KEY=your_gemini_key_here

4. Configure agents and tasks:
   • Edit config/agents.yaml and config/tasks.yaml to customize agent behavior and tasks.

Troubleshooting

• Ensure your API keys are valid and have access to the selected LLM provider.
• If you see authentication errors, double-check your .env file and restart your terminal.
• For C++ build errors on Windows, install Microsoft C++ Build Tools.

License

MIT