The GPT Content Reviewer is a powerful and versatile application that integrates OpenAI's GPT-3.5-turbo model to provide various functionalities such as general chat, task processing, file processing, eye-tracking data review, chat-based planning, AI wrangling, and AI rerouting. This application is built using Flask, a lightweight web framework for Python.
To install and run the GPT Content Reviewer, follow these steps:
- Clone the repository:
git clone https://github.com/your_username/gpt-content-reviewer.git
- Navigate to the project directory:
cd gpt-content-reviewer
- Install the required dependencies:
pip install -r requirements.txt
- Set your OpenAI API key as an environment variable:
export OPENAI_API_KEY="your_api_key_here"
- Run the application:
python app.py
The application should now be running on http://localhost:5000.
The general chat functionality allows users to engage in a conversation with the GPT-3.5-turbo model. Simply type your message into the input field and press "Send" to receive a response from the AI.
The task processing functionality enables users to provide a task description, and the application will recommend the best approach, research related topics, and generate code to complete the task. The output will be presented in a human-readable format.
The file processing functionality allows users to process .docx and .xlsx files. The application will rewrite the content of the files while retaining their original meaning. The rewritten files will be saved with a "_rewritten" suffix.
The eye-tracking data review functionality enables users to review eye-tracking data stored in a CSV file. The application will provide a review of the data and return the result in a JSON format.
The chat-based planning functionality allows users to engage in a conversation with the GPT-3.5-turbo model to plan a specific task or project. Users can continue the conversation until they decide to stop, and the application will provide a summary of the planning session.
The AI wrangling functionality enables users to provide a message, and the application will generate a "wrangled" version of the message using the GPT-3.5-turbo model.
The AI rerouting functionality allows users to provide a message and a target model, and the application will reroute the message to the specified target model, generating a response accordingly.
The GPT Content Reviewer provides the following API endpoints:
/: Render the main page of the application./retro_style.css: Serve the CSS file for the application./AI_chat: Process user input and return a chat response from the GPT-3.5-turbo model./api/process_folder: Process a folder containing.docxand.xlsxfiles./api/process_csv: Process a CSV file containing eye-tracking data.
Contributions are welcome! Please feel free to submit a pull request or create an issue to discuss any changes or improvements.
This project is licensed under the MIT License. See the LICENSE file for details.