Will ChatGPT ever be given ability to read our Repositories code in GitHub? As access to this would be very handy
Do you have ChatGPT Plus?
Are you using the ChatGPT Plus Plugin store?
These plugins are in the store for GitHub.
Do you want me to demo one? Need details on what you would like to see. It will not be a video just a few screenshots.
For what is in the ChatGPT Plugin Store I like this 3rd party website.
I did not know of them plugins thanks for pointing that out.
I have found one problem, that is the github plugins do not work if you want to use code interrupter. you can only use it for the plugins, this makes it a real pain. i say this as I’m tired of the goldfish memory of this chatgpt, i just seem to have to keep repeating myself.
I can not link code interrupter to github. I can not say to chatgpt here is my old code keep that info aside as a log till we add updates and then we update the log with the new code. be nice if it also displayed the old and new code showing the changers we are making or have made.
here it the problems i run into coding with chatgpt :
I apologize for any confusion, but I need more specific information to assist you effectively. Each of the files mentioned (overlay.py
, custom_dialog.py
, text_overlay_dialog.py
, and image_overlay_dialog.py
) corresponds to different modules or components of your PyQt5 application, each potentially having a different purpose and functionality.
The overlay.py
file, for example, seems to handle the overlay component of your PyQt5 application. The custom_dialog.py
file likely contains the definition and functionality for a custom dialog box in your application.
Without specific information about what each of these modules is intended to do, it’s difficult to provide accurate guidance.
Could you provide more details about the intended functionality of these files or the specific issues you’re encountering? This will allow me to give more directed and helpful advice.
my reply : here are the files you requested. code overlay.py (files and code added )
Chatgpt reply: From your code, it looks like you have already implemented the necessary classes for your PyQt5 application: Overlay
, CustomDialog
, TextOverlayDialog
, and ImageOverlayDialog
. These classes seem to be correctly implemented based on the requirements you’ve provided, so I’m not sure what exactly needs to be fixed.
The error you are facing could be due to how you’re integrating these classes into your main_window.py
. Ensure that you are correctly importing these classes and using them within your MainWindow
class.
Here’s how you might import these classes at the beginning of your main_window.py
:
pythonCopy code
* *from overlay import Overlay* *from custom_dialog import CustomDialog* *from text_overlay_dialog import TextOverlayDialog* *from image_overlay_dialog import ImageOverlayDialog* *
And here’s how you might use these classes within your MainWindow
class:
pythonCopy code
```
self.overlay1 = Overlay(self)
self.overlay2 = Overlay(self)
self.imageOverlayDialog = ImageOverlayDialog(self, self.overlay2)
self.textOverlayDialog = TextOverlayDialog(self, self.overlay1)
```
If you are still having trouble, please provide more specific information about the errors or problems you are encountering. This might include error messages you’re seeing, unexpected behaviour from your application, or features that aren’t working as expected. With more specific information, I can provide more targeted help.
Also, please note that I am an AI language model developed by OpenAI and I do not have the capability to run or test your PyQt5 application directly. I recommend running your application locally and observing any errors or issues that occur, which you can then ask me for help with.
Problem i have is all this because the code has no memory, and now because of its goldfish memory losses track of the functions, so it can not keep track of task longer then 2 replies, really makes this program frustrating to work with.
You do realize that you have gone off topic in your own topic and started a new question, which because it is a reply in a topic with an unrelated title will not be seen by many.
To respond to your question, ChatGPT was never designed to be used that way. Consider another solution.
You do realize that you have gone off topic in your own topic and started a new question, which because it is a reply in a topic with an unrelated title will not be seen by many.
I have found one problem, that is the github plugins do not work if you want to use code interrupter.
How is that going off topic, when my intention was to use code interrupter with the plugin but this is not possible. and i made a statement that chatgpt has a memory of a goldfish because i have to feed all my code in then it losses track what its doing instead of just reading from my repository. so i’m for ever repeating my self and im getting tired of it.
The topic is " Will ChatGPT ever be given ability to read our Repositories code in GitHub?"
There is no mention of Code Interpreter.
That was not noted in the topic title or the first post. We do not read minds.
The info was given to you as friendly advice. Just start a new topic.
Will ChatGPT ever be given ability to read our Repositories code in GitHub?
yes to internal functions like code interrupter the plugin suggested are useless to me if they do not work with code interrupter i just assumed you would know this if I’m wanting to code! as the normal coder is as good as 3.5 and i pay for gpt4 so i want the functions to work with github.
This is what Github Copilot is designed to do (to look at github)
So not sure if they would add it to chatGPT
(I love copilot for coding)
Consider it a modifier for plugins being one solution…
- Will ChatGPT ever be given ability to read our Repositories code in GitHub,
- by using plugins,
- while code interpreter mode is enabled?
And then one considers the amplification of possible bot attacks on the internet by executing AI-written and monitored code in combination with the internet access a plugin gives.
I’m over the fear mongering of people about chatgpt, I believe in progress and the positive impact technology can have on our world. It bothers me that some companies seem more concerned about safeguarding their financial interests than embracing the potential benefits of AI. Instead of holding back, I am determined to contribute to the improvement of our society.
In this fast-paced world, it’s essential to keep up with innovation; otherwise, we risk falling behind. ChatGPT, being a promising tool, must continue to evolve and stay ahead. Let’s not be the first to lag behind while others surge forward, leaving us years behind in just a few months.
I understand the importance of internet safety, and I support measures to counter misuse of AI but there are save guards to stop this behaviour on ChatGPT. If people want to jailbreak a LLM it will be done on a (torrent share LLM that runs LLaMA 1 and 2) where you all hold 1 TB of data each and share the new models on the fly.
Platforms like GitHub could implement terms to prevent the development of harmful tools that may exploit from ChatGPT.
As the saying goes, “The only thing to fear is fear itself.” We shouldn’t let fear hinder our progress. Instead, let’s focus on using technology responsibly to create a better future.