code-review-graph

Code-review-graph extracts the local priority structural map of your codebase, allowing AI coding tools to read only the relevant context. Using Tree-sitter, it tracks changes incrementally and provides sensitive context to your assistant through the MCP protocol. It aims to reduce token consumption for large warehouses.

CO

tirth8205 · AI Coding Assistants

STARS32.5K
FORKS3.0K
COMMITS1.0K
CONTRIBUTORS120
VERSIONv2.3.8
UPDATED2 days ago
PythonTree-sitterMCPCode IntelligenceMIT

Common commercial products in this category: GitHub Copilot, Cursor, Tabnine

What can it do?
  • With Tree-sitter, you can create a permanent structural map of your codebase.
  • You can automatically install the MCP configuration for supported platforms.
  • By giving the AI assistant sensitive context, you can only make the required code parts read.
User reviews
  • • NeutralHacker News

    Code-review-graph: persistent code graph that cuts Claude Code token usage

    “Hi HN I'm Tirth. I built code-review-graph because I got tired of watching Claude Code re-read my entire codebase on every single task. When you ask Claude Code to review a commit or add a feature, it reads files to understand the codebase. On a small project that's fine. On…”

Verified against README · Oct 26, 2026 · source

My note

Alternative pages

More in: AI Coding Assistants