kodus-ai
AI-powered code review tool that lets developers choose and control the underlying LLM.
freemium1.4k github stars
kodus-ai 简介
Kodus AI is a coding assistant focused on automated code review. It integrates into development workflows to analyze pull requests and suggest improvements.
The tool emphasizes transparency and cost control by allowing users to select from multiple language models—including open-source and proprietary options—and configure inference parameters like temperature and max tokens.
It supports GitHub integration and is designed for teams that want granular control over AI behavior, model selection, and budgeting per review.
核心功能
- GitHub-native pull request code review automation
- Support for multiple LLMs including local and cloud-hosted models
- Customizable review criteria and severity thresholds
- Cost tracking per review based on model and token usage
- Configurable review templates for team-specific standards
- Self-hosted option for enterprise security requirements
优点 & 缺点
优点
- • Full control over model choice and inference settings
- • Transparent cost estimation before review execution
- • Supports both cloud and self-hosted deployments
缺点
- • No public documentation on free tier limits or feature restrictions
- • Requires technical setup for model configuration and integration