OSS Alternative - Discover Top Open Source Alternatives to Popular Software

tirth8205/code-review-graph

Optimizes AI coding tools by building a local knowledge graph of your codebase, significantly reducing token usage for code reviews and daily coding tasks.

Core Features

Builds a structural knowledge graph of your codebase using Tree-sitter.

Reduces AI token usage by providing precise context, leading to up to 49x savings.

Integrates with various AI coding platforms (e.g., Claude Code, Cursor, Codex) via MCP.

Automatically updates the graph on file edits and Git commits.

Auto-detects and configures supported AI coding tools with a single 'install' command.

Quick Start

pip install code-review-graph && code-review-graph install

Detailed Introduction

Code-review-graph addresses the inefficiency of AI coding tools re-reading entire codebases for every task. It constructs a persistent, local knowledge graph of your project using Tree-sitter, mapping functions, classes, and their relationships. This graph enables AI assistants to receive only the most relevant context via the Model Context Protocol (MCP), drastically cutting down token consumption for code reviews and daily development. It offers seamless integration with popular AI platforms, automating setup and ensuring the graph remains up-to-date with codebase changes, thereby enhancing developer productivity and reducing operational costs.