Command line tool to leverage OpenAI's API for various marketing tasks.
- Identify Niche Problem and Solution
- Define Target Audience and Create Avatar
- Write a Sales Letter in Story Form Using PAS Framework
- Create a Landing Page for the Product or Service
- Python 3.6 or higher
- OpenAI API key(see OpenAI API for more information)
- OpenAI Assistant ID token(see OpenAI API for more information)
- Clone the repository:
git clone https://github.com/rp42dev/OpenAI-Marketing-Assistant.git
- Install dependencies:
pip install -r requirements.txt
- Set up environment variables:
Create a .env
file in the root directory of the project and add the following environment variables but make sure to replace the values with your own.
- Set up configuration file:
Create a config.json
file in the root directory of the project and add the following configuration, use the example below as a guide.
- Run the application:
python main.py
OPENAI_API_TOKEN=sk-proj-1234567890...
ASSISTANT_ID_TOKEN=asst_1234567890...
{
"task_groups": {
"Market Research and Targeting": {
"Identify Niche Problem and Solution": {
"instructions": "Identify a specific problem within a niche and propose a solution that product or service addresses. Provide a brief description of both the problem and how the product or service can solve it."
},
"Define Target Audience and Create Avatar": {
"instructions": "Define the target audience and create a detailed avatar. Include demographics, psychographics, and any other relevant details."
}
},
"Marketing and Promotion": {
"Write a Sales Letter in Story Form Using PAS Framework": {
"instructions": "Write a sales letter in the form of a story using the Problem-Agitate-Solution (PAS) framework. The letter should address the target audience's pain points, agitate the problem, and present the product or service as the solution."
},
"Create a Landing Page for the Product or Service": {
"instructions": "Create a landing page for the product or service that includes a headline, subheadline, call-to-action, and any other relevant information. The landing page should be designed to convert visitors into leads or customers."
}
}
},
"function_groups": {
"corrections": {
"Task Correction": {
"instructions": "Make corrections to the previous response based on user input."
}
}
}
}
Contributions are welcome! If you have any ideas, suggestions, or bug fixes, please feel free to open an issue or submit a pull request.
This project is licensed under the MIT License.