✨ Feature Request: AI-Powered GitHub Issue Generator (using o3-mini)

[DonnieBLT]

We propose adding a new feature to the BLT website that leverages OpenAI’s o3-mini model to automatically generate detailed GitHub issues from a simple user-provided description. Users will be able to input a brief summary of a feature, bug, issue, or task, and the system will use the o3-mini model to produce a fully detailed GitHub issue complete with AI agent instructions for further development or resolution.

Feature Details:
• User Input:
• A text input area where users can describe a feature, bug, issue, or task in plain language.
• AI Integration:
• Integration with the OpenAI API using the o3-mini model.
• The model processes the user’s input and generates a detailed GitHub issue that includes:
• A comprehensive description.
• Steps to reproduce (if applicable).
• Suggested tasks or subtasks.
• Detailed AI agent instructions for potential automation or further processing.
• Output Display:
• The generated GitHub issue will be presented on the website for review.
• Users will have the option to directly copy or submit the issue to a GitHub repository via an authenticated integration.
• Configuration:
• API keys and model parameters should be managed securely using environment variables.
• The prompt template used by the o3-mini model for generating issue details should be easily configurable.

Technical Requirements:
• Frontend:
• A simple, user-friendly form component on the BLT website.
• A display area for the AI-generated GitHub issue preview.
• Backend:
• A new Django endpoint that:
• Accepts user input.
• Calls the OpenAI API using the o3-mini model.
• Returns the generated issue text.
• Robust error handling for API failures or unexpected input.
• Testing:
• Unit tests for the API integration with the o3-mini model.
• Integration tests ensuring the generated issues meet the expected format and content.
• Documentation:
• Clear instructions on configuring the OpenAI API integration and adjusting the prompt template.
• A user guide explaining how to use the AI-powered issue generator feature.

AI Coding Agent Instructions:

To assist in building this feature, use the following prompt for your AI coding agent:

You are tasked with building a Django web application feature called "AI-Powered GitHub Issue Generator" for the BLT website. The feature should allow a user to input a short description of a feature, bug, issue, or task, and then use the OpenAI API with the o3-mini model to generate a detailed GitHub issue complete with a comprehensive description, reproduction steps, subtasks, and detailed AI agent instructions.

Requirements:

• Create a Django app (e.g., named "issue_generator") with the following:
  • A model (if necessary) to store user submissions and generated issues.
  • A view that handles a form submission (using a Django form) from the homepage where the user enters their description.
  • A Django view (or API endpoint) that takes the user input, calls the OpenAI API (using the o3-mini model), and returns the generated GitHub issue.
  • Template(s) to display the input form and the output preview of the generated issue.
  • Error handling to manage API failures gracefully.
  • Unit tests for the API integration and view logic.

Additional notes:

• All API keys and model parameters should be read from environment variables.
• Ensure that the code is modular so that the prompt template for the OpenAI API can be easily updated.
• Provide comments in the code to explain the logic and data flow.
• Follow best practices for Django project structure.

Generate the scaffolding code, including models (if any), views, forms, templates, and unit tests, for this feature.

Acceptance Criteria:
1. Functionality:
• A user can enter a brief description on the BLT website.
• The system sends this description to the OpenAI API using the o3-mini model and retrieves a detailed GitHub issue.
• The generated issue includes a comprehensive description, reproduction steps (if applicable), subtasks, and clear AI agent instructions.
2. Usability:
• The feature is integrated with a clean, intuitive UI.
• Users can easily review and copy/submit the generated issue.
3. Reliability:
• The API integration handles errors gracefully and provides informative feedback to the user.
• All configurations (API keys, model parameters, prompt templates) are securely managed.
4. Testing & Documentation:
• Comprehensive unit and integration tests cover the new feature.
• Documentation is updated with setup instructions, configuration details, and usage examples.

Additional Considerations:
• Monitor potential rate limits or usage costs associated with the OpenAI API and implement safeguards as needed.
• Consider future enhancements such as saving past generated issues, allowing in-browser edits before submission, or providing multiple generation options based on user preferences.

[krrish-sehgal]

/assign

[github-actions]

Hello @krrish-sehgal! You've been assigned to OWASP-BLT/BLT issue #3311. You have 24 hours to complete a pull request.

[github-actions]

⏰ This issue has been automatically unassigned due to 24 hours of inactivity.
The issue is now available for anyone to work on again.

[krrish-sehgal]

/assign

[github-actions]

Hello @krrish-sehgal! You've been assigned to OWASP-BLT/BLT issue #3311. You have 24 hours to complete a pull request.

[krrish-sehgal]

@DonnieBLT sir , in the requirements section , its mentioned to "Create a Django app (e.g., named "issue_generator")" , I wanted to double check if this is what needs to be done, as it can be implemented under the website app only.

[DonnieBLT]

No separate app. 👍 only the website app with organized files

[github-actions]

⏰ This issue has been automatically unassigned due to 24 hours of inactivity.
The issue is now available for anyone to work on again.

[github-actions]

⏰ This issue has been automatically unassigned due to 24 hours of inactivity.
The issue is now available for anyone to work on again.

[krrish-sehgal]

@DonnieBLT sir, I have been trying to figure out a lot , about making the exsisting django allauth functionality work with github.

I have went through numerous resources including stackoverflow and documentations to figure out , where does the allauth token is stored , everywhere it was suggested that it automatically stores in Social token model, but this access token is not getting stored. Moreover all auth is offering a really bad UX , and each time we will have to go to the dummy allauth page , to login.

Can i make customised functions to our usecase and remove allauth for github login, it will allow us to have more control.

[DonnieBLT]

Allauth was working before. I would suggest first removing the custom code and adding it back one by one so we can still use it and keep a standardized code

[krrish-sehgal]

Sure sir , I'll make it work .

[krrish-sehgal]

@DonnieBLT sir , figured out the configs after taking some help from the django community.
Its working now. Can you please review the PR?

Thanks