What I built
I have built AdaGPT, an AI-powered GitHub Action that generates helpful responses for Issues and Pull Requests. It's like having a conversation with ChatGPT, but without actually leaving GitHub and available to everyone.
All you have to do is mention @AdaGPT in your issue or comments and she will respond to you with a helpful and charming reply.
Category Submission:
Wacky Wildcards
App Link:
Screenshots
Here are some screenshots of Issues and Pull Requests
Explain Pull Requests
Rock, Paper, Scissors!
With a little luck, you might convince AdaGPT to play a Rock, Paper, Scissors with you!
Tic, Tac, Toe
Description
AdaGPT is a GitHub Action that responds to issues and pull requests if they mention @AdaGPT
.
AdaGPT takes into account the description of the issue as well as all existing comments. For pull requests, AdaGPT also has access to the Git diff the changes.
Link to Source Code
Permissive License
MIT
Background (What made you decide to build this particular app? What inspired you?)
Trying out AI has been on my list of side projects for a long time. This hackathon finally gave me the motivation to tackle it. The idea for this GitHub activity came to me while using ChatGPT. A conversation using ChatGPT is similar to a discussion in an issue or pull request. However, on GitHub, multiple people can participate in the discussion and the messages are visible to everyone. Therefore, it sounded like an exciting idea to simply include ChatGPT in conversations on GitHub and make it available to everyone.
How I built it (How did you utilize GitHub Actions or GitHub Codespaces? Did you learn something new along the way? Pick up a new skill?)
The GitHub Action was implemented with TypeScript using the GitHub Actions Toolkit and the OpenAI Node.js Library.
The Action is activated on issues, pull requests, and comments, if the description or comment contains @AdaGPT
.
The relevant issue number is determined from the event payload and doesn't need to be passed separately. In case of a pull request, the git diff of all changes is retrieved from the GitHub API. Furthermore, if the Action was triggered by a comment, all existing previous comments are read from the API.
Finally, a corresponding prompt is generated from all information (issue, pull request diff, comments) and the request is sent to OpenAI. The response of the request is added as a new comment to the issue.
Top comments (7)
Whilst it's nice that you've recognised Ada, Countess of Lovelace, in naming your application, the fact that there is already an Ada programming language with numerous tools whose names are something like 'AdaXXX' etc could make people think you're actually programming your application in Ada.
Yes, at first glance some people might think that it is a GPT implementation in Ada. On the other hand, I think that Ada as a programming language is not that common nowadays, at least not on GitHub. In fact, the first GPT3 model of OpenAI was called Ada. I think (and hope) most people will think of the person Ada when they hear
AdaGPT
.Really impressive
great tool
Great use case!
Great use case for the GPT models 👍🏻 Ada helps with documentation while engineers have more time tackling new issues
Rock paper scissors is kinda failure for ChatGPT as I can see here lol