Code Expert System
Createdtomsiwik
Simulate code reviews from expert personas using AI and a knowledge graph.
About
This Python-based code review system leverages the Model Context Protocol (MCP) to provide expert feedback on your code changes. It simulates code reviews from well-known experts like Martin Fowler and Robert C. Martin (Uncle Bob), offering insights based on their established principles. The system uses Ollama for AI-powered reviews and stores code, reviews, and relationships in a knowledge graph, enabling a rich understanding of your codebase and its evolution.
Key Features
- Code review based on Martin Fowler's refactoring principles
- Knowledge graph storage of code, reviews, and relationships
- Server-side Event (SSE) support for web integration
- Code review based on Robert C. Martin's Clean Code principles
- Integration with Ollama for AI-powered reviews
Use Cases
- Automated code review and suggestions
- Integrating code review into Cursor IDE
- Learning best practices from simulated experts