Convert your ChatGPT conversations to Markdown

Overview

This project consists of a Python script, convert.py, which is designed to process conversation data from ChatGPT and convert it into Markdown format. The script takes a JSON file containing conversation data and outputs a series of Markdown files, each representing a conversation thread.

You can export your data by following this documenation below

https://help.openai.com/en/articles/7260999-how-do-i-export-my-chatgpt-history-and-data

Getting Started

Prerequisites

• Python 3.x
• Access to the command line or terminal

Installation

No additional libraries are required for this script, as it only uses modules from the Python Standard Library.

Usage

1. Prepare Your Data: Ensure your conversation data is in a JSON format compatible with the script.

2. Run the Script: Use the command line to navigate to the directory containing convert.py and run the following command:

   python convert.py [input_file] [output_dir] [--use-date-folders]

   • Replace [input_file] with the path to your JSON file containing the conversation data.
   • Replace [output_dir] with the path to the directory where you want the Markdown files to be saved.
   • Include --use-date-folders if you want to store the Markdown files in folders named by the date of the conversation in ISO format. Examples:

   # Without date-based folders
   python convert.py conversations.json output
   
   # With date-based folders
   python convert.py conversations.json output --use-date-folders

3. Check Output: After running the script, the Markdown files will be saved in the specified output directory, with each file representing a conversation from the input data.If --use-date-folders is used, each conversation's Markdown file will be stored in a folder named after the conversation's date in ISO format (e.g., 2023-01-01).

File Structure

• convert.py: The main script that processes the JSON data and converts it to Markdown format.
• README.md: This file, containing instructions and information about the project.

License

This project is open source and available under MIT.