CodeSage

🦀 AI-powered Code Review & Refactoring Tool built with Rust

Overview

CodeSage is an intelligent code analysis and refactoring tool that combines the power of static analysis, semantic understanding, and AI to help developers write better code.

Key Features

• 🔍 Smart Code Review: Go beyond syntax - understand logic, edge cases, and best practices
• ♻️ Intelligent Refactoring: AI-driven suggestions for improving code structure and maintainability
• 📊 Technical Debt Tracking: Quantify and visualize code quality over time
• 🔒 Security Scanning: Identify vulnerabilities and security issues
• 🚀 High Performance: Built with Rust for speed and reliability
• 🔌 Multiple Interfaces: CLI, VS Code extension, Git hooks, and web dashboard
• 🏠 Privacy-First: Support for local deployment and offline models

Supported Languages

• Rust
• JavaScript / TypeScript
• Python
• Go
• Java
• C/C++
• More coming soon...

Quick Start

# Install via cargo (coming soon)
cargo install codesage

# Review a file
codesage review src/main.rs

# Review entire project
codesage review . --recursive

# Interactive refactoring
codesage refactor src/utils.rs --interactive

# Generate technical debt report
codesage debt --output-html report.html

Project Status

🚧 Early Development - This project is currently in active development.

Roadmap

• Core architecture and AST parsing (tree-sitter)
• Basic static analysis engine
• AI integration (Claude/GPT/Local LLM)
• CLI interface
• VS Code extension
• Web dashboard
• CI/CD integrations

Architecture

codesage/
├── crates/
│   ├── codesage-core/       # Core analysis engine
│   ├── codesage-parser/     # Multi-language code parsing
│   ├── codesage-analyzer/   # Static & semantic analysis
│   ├── codesage-ai/         # AI/LLM integration
│   ├── codesage-refactor/   # Refactoring engine
│   └── codesage-cli/        # Command-line interface
├── docs/                    # Documentation
├── examples/                # Usage examples
└── tests/                   # Integration tests

Contributing

Contributions are welcome! Please read our Contributing Guide first.

License

MIT License - see LICENSE file for details

Acknowledgments

Built with:

• tree-sitter - Incremental parsing system
• Rust - Performance and safety
• AI models from Anthropic, OpenAI, and open-source community

---

Note: This is an independent open-source project and is not affiliated with Anthropic's Claude Code.