🧠 Ad Campaign Creation Agent

An LLM-driven agent that converts campaign briefs into structured ad campaign plans.

---

🚀 Features

• Converts natural language campaign briefs into machine-readable JSON plans
• Generates multiple ad creative variants with justifications
• Performs automated consistency checks
• Supports mock LLM responses for testing
• Includes example briefs and outputs

---

⚙️ Setup

1. Clone the repository

   git clone https://github.com/yourusername/campaign-agent.git
   cd campaign-agent

2. Create a virtual environment

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

3. Install dependencies

   pip install -r requirements.txt

4. Configure environment variables

   cp .env.example .env

   Then add your OpenAI API key and model settings inside .env.

---

🖥️ Usage

Streamlit UI

Run the web interface:

streamlit run ui/app.py

The UI provides a user-friendly interface for:

• Creating and editing campaign briefs
• Generating campaign plans
• Viewing detailed results
• Downloading campaign plans as JSON

---

Command Line

Process a campaign brief:

python -m src.agent examples/brief1.json

---

Python API

from src.agent import CampaignAgent

# Initialize agent
agent = CampaignAgent()

# Process brief from file
campaign = agent.process_brief_from_file("examples/brief1.json")

# Or process brief directly
brief = {
    "campaign_id": "cmp_2025_09_01",
    "goal": "increase trial signups",
    # ... rest of brief ...
}

campaign = agent.process_brief(brief)

---

🧪 Testing

Run tests with mock LLM responses:

pytest tests/

---

📁 Project Structure

campaign-agent/
├── src/                     # Main source code
│   ├── agent.py             # Campaign agent implementation
│   ├── models/              # Data models
│   ├── prompts/             # LLM prompts
│   ├── validators/          # Validation logic
│   └── utils/               # Utility functions
├── tests/                   # Test files
├── examples/                # Example briefs and outputs
├── requirements.txt         # Python dependencies
└── .env.example             # Environment variables template

---

🔑 Environment Variables

Variable	Description	Default
OPENAI_API_KEY	Your OpenAI API key	—
LLM_MODEL	Model to use	gpt-4
TEMPERATURE	Model temperature	0.7
MAX_TOKENS	Maximum tokens per response	2000
MOCK_LLM	Use mock responses for testing	false

---

🧩 Limitations and Future Improvements

1. Limited grounding: Currently doesn’t use a knowledge base for product facts or past metrics
2. Basic validation: Could add more sophisticated checks for ad copy quality
3. No performance prediction: Future versions could include ML-based creative scoring

---

📜 License

MIT License

---

Project: radiac.ai_assignment