name: 'Usage of code-review-gpt GitHub Action'
on:
pull_request:
branches: [main]
jobs:
deploy:
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v4
- name: Code Review GPT
uses: mattzcarey/code-review-[email protected]-alpha
with:
OPENAI_API_KEY: ${{ secrets.OPENAI_API_KEY }}
MODEL: 'gpt-3.5-turbo'
GITHUB_TOKEN: ${{ github.token }}
Code Review GPT
Your personal code reviewer powered by LLMs (OpenAI GPT-3.5/4) & Embeddings ⚡️ Improve code quality and catch bugs before you break production 🚀 Lives in your Github/GitLab/Azure DevOps CI
What is Code Review GPT?
Code Review GPT seems to me to be a single giant leap, built on the huge general language model that has been making rounds lately. It seems that this tool built on large language models is going to be able to automate the process of reviewing source code changes and, in effect, be able to detect potential issues before these issues manifest in the production environment. Even when Code Review GPT promises to make the code review process efficient by detecting most of the common issues that arise during coding, it is pertinent to remember that this is a product in alpha release and should be used as a supplementary tool for human code review, more so in professional environments where reliability is of prime importance.
Tips for Effective Usage
- Verify Suggestions: Always verify the AI’s suggestions manually to ensure they are appropriate and do not introduce new issues.
- Use in Development: Initially integrate Code Review GPT in development environments or non-critical paths to assess its effectiveness and reliability.
- Feedback Loop: Provide feedback on the tool’s performance and suggestions to help improve its accuracy and usefulness.