Appearance
๐ Official GitHub Integration โ
The LuxurAI GitHub Connector connects your GitHub repositories directly into your LuxurAI conversational workflow. Instead of copy-pasting code back and forth, you can grant the AI access to read repositories, explain architectures, draft pull requests, and commit code directly to branches.
๐ Setting Up GitHub Connection โ
- Open the Connectors Modal from the chat interface.
- Under the Discover tab, find the GitHub integration card.
- Click Connect with GitHub.
- You will be redirected to GitHub's secure OAuth consent screen.
- Choose which repositories (all repositories or select specific projects) LuxurAI is permitted to access.
- Once authorized, you are returned to LuxurAI with your GitHub workspace fully linked.
๐ป What You Can Do in the UI โ
1. Repository Grounded Conversations โ
Anchor your chat session to any connected repository:
- Select your target repository from the active context dropdown.
- Ask questions like:
- "Where is authentication middleware defined in this repo?"
- "How does the payment webhook verify signatures?"
- "Audit this codebase for potential memory leaks or missing rate limiters."
2. Multi-File Code Synthesis & Live Commits โ
LuxurAI doesn't just suggest snippets โ it can write and commit code directly:
- Request a feature implementation (e.g. "Add a dark mode toggle to the navbar").
- LuxurAI synthesizes the code, presents an interactive diff preview, and lets you review the modifications.
- Click Commit to Branch or Create Pull Request directly from the chat window.
3. Automated PR & Issue Summaries โ
- Point LuxurAI to an open issue or Pull Request.
- Get a succinct architectural overview, breakdown of changed files, and automated regression risk assessment.
๐ก๏ธ Security & Permissions โ
- Fine-Grained Scopes: LuxurAI only requests the minimal GitHub permissions needed for repo reading and commit synthesis.
- Revoke at Any Time: You can disconnect your GitHub account or revoke repository permissions instantly via the Yours tab in Connectors or from your GitHub Account Settings.
- Zero Storage of Private Code: Your source files are processed transiently in memory for prompt context and are never retained or used to train public models.
