Reverse Engineered ChatGPT by OpenAI. Extensible for chatbots etc.
Note: This is a proof of concept. The goal here is to be able to write bots using the API.
- Scientific knowledge
- Has memory. It remembers the chat history and context
- No moderation
- Programmable
pip3 install revChatGPT
Go to https://chat.openai.com/chat and log in or sign up
- Open console with
F12 - Go to Network tab in console
- Find session request (Might need refresh)
- Copy accessToken value to config.json.example as Authorization
- Save as config.json (In current active directory)
python3 -m revChatGPT
Remember to press enter twice to send the message. This allows for multi-line input.
You can pass the output through a processor using command line arguments.
Example:
Use python3 -m revChatGPT say to make a Mac speak the output
pip3 install revChatGPT
from revChatGPT.revChatGPT import Chatbot
import json
# Get your config in JSON
config = {
"Authorization": "<Your Bearer Token Here>"
}
chatbot = Chatbot(config, conversation_id=None)
prompt = "<Some prompt>"
response = chatbot.get_chat_response(prompt)
print(response["message"])
print(response["conversation_id"])
print(response["parent_id"])This can be imported to projects for bots and much more. You can have multiple independent conversations by keeping track of the conversation_id.
- Access token expires in one hour