An intelligent AI-powered GitHub issue triage agent that automatically classifies, labels, and comments on new issues using Python, Flask, and GPT-4.

• AI-Powered Classification: Uses GPT-4 Turbo for intelligent issue categorization.
• Automatic Labeling: Applies appropriate labels based on AI analysis.
• Smart Comments: Generates detailed triage explanations.
• Secure Webhooks: Validates GitHub webhook signatures for secure communication.
• Real-time Processing: Instantly triages new issues as they are created.
• Extensible: Built with a clear structure using Flask and a dedicated triage agent class.

1. Webhook Reception: A Flask server listens for webhook events from a GitHub repository whenever a new issue is opened.
2. Signature Verification: The server verifies the webhook payload's signature to ensure it originated from GitHub.
3. Issue Processing: The issue details are passed to the TriageAgent.
4. AI Classification: The agent sends the issue title and body to the OpenAI API (or a compatible service) to determine the issue's nature (e.g., bug, feature request).
5. Labeling and Commenting: Based on the AI's response, the agent uses the GitHub API to add the appropriate labels and post a comment on the issue.

1. Python 3.9+ and pip.
2. GitHub Personal Access Token with repo scope.
3. OpenAI API Key.
4. A GitHub Repository to install the agent on.
5. A Webhook Secret to secure the communication between GitHub and your agent.

1. Clone the repository:

   git clone <this-repo>
   cd github-triage-agent

2. Create a virtual environment and install dependencies:

   python -m venv venv
   source venv/bin/activate  # On Windows use `venv\Scripts\activate`
   pip install -r requirements.txt

3. Configure environment variables: Create a .env file in the project root and add the following variables. You can copy the example below.

   # .env
   GITHUB_WEBHOOK_SECRET=your_webhook_secret
   GITHUB_TOKEN=your_github_personal_access_token
   GITHUB_REPO_OWNER=owner_of_the_repo
   GITHUB_REPO_NAME=name_of_the_repo
   OPENAI_API_KEY=your_openai_api_key

4. Set up the GitHub Webhook:

   • Go to your target GitHub repository's Settings > Webhooks.
   • Click Add webhook.
   • Payload URL: Set this to the public URL where your Flask app will be running (e.g., https://your-domain.com/webhook). For local development, you can use a tool like ngrok to expose your local server to the internet.
   • Content type: application/json.
   • Secret: Use the same secret you defined in your .env file.
   • Which events would you like to trigger this webhook?: Select "Issues".
   • Make sure the webhook is active.

5. Create required labels in your GitHub repository. The agent will try to apply these labels, so they should exist:

   • bug
   • feature-request
   • documentation
   • question
   • enhancement

# Activate the virtual environment
source venv/bin/activate

# Run the Flask application
python main.py

The server will start, typically on http://127.0.0.1:5000.

MIT License