English README | 中文 README

Description 📚

A project using LLM to empower Code Review 💪🏻 ((LLM not only includes GPT 🎁))

What are the characteristics of the project? ✨

🐶 We customize it for gitlab

🐱 We combine the powers of GPT 🚀

🦊 We're going to access privatization LLM to solve code security issues

🦁 We will always be aware of the latest developments of code review into the project

Project structure 🚗

Results preview 🌈

1. Gitlab Code Review for Merge Request

2. Dingding-code-review message

3. Integration of More Large Models

1. Custom models can be quickly integrated by implementing the project interface. For specific configuration methods, refer to config.md.
2. The project supports multiple models through UnionLLM, which is compatible with LiteLLM. The default supported models are listed in the table below.

OpenAI	Azure	AWS - SageMaker	AWS - Bedrock
Google - Vertex_AI	Google - Palm	Google AI Studio - Gemini	Mistral AI API
Cloudflare AI Workers	Cohere	Anthropic	Empower
Huggingface	Replicate	Together_AI	OpenRouter
AI21	Baseten	Vllm	NLP_Cloud
Aleph Alpha	Petals	Ollama	Deepinfra
Perplexity-AI	Groq AI	DeepSeek	Anyscale
IBM - Watsonx.ai	Voyage AI	Xinference [Xorbits Inference]	FriendliAI
Galadriel	智谱AI	月之暗面 Moonshot	百度文心一言
阿里巴巴通义千问	MiniMax	讯飞星火	百川智能
昆仑天工	零一万物	阶跃星辰	字节豆包
DeepSeek	More

# Usage 📖

install

Run by Source code

1.clone repo

git clone git@github.com:mimo-x/Code-Review-GPT-Gitlab.git

2.install requirements

pip install -r requirements.txt

3.modify config file

vim config/config.py

4.run

python3 app.py

5.config gitlab webhook

When filling in the Webhook URL, please add /git/webhook after the domain name. For example: http://example.com/git/webhook

6.**make a mr 🎉**

Docker

todo dockerfile

Todo List 📌

• ✅ Use GPT to Code Review
• Try to access the privatization LLM to Address code security issues
• Enable more trigger methods
  • ✅ Merge Request
  • for every commit
  • for tag
• FeiShu(飞书) Code Review message notification
• Static code analysis to Analyze the risk level of modifying code

Communication 👨‍👨‍👦‍👦

👏🏻You're welcome to ask us questions（issue，pr）, Thank you for your star ⭐️

📮 Email：mixuxin@163.com

👨‍👨‍👦‍👦 This is our WeChat group

reference 📚

• (字节)基于大模型 + 知识库的 Code Review 实践
• (美团)代码变更风险可视化系统建设与实践

License 📑

This project is licensed under the MIT License.