• Picks
  • Discover
  • Alternatives
  • About

© 2026 Willet

  • Newsletter
  • About
  • FAQ
  • Terms
  • Privacy & Cookies
Join the Discord (opens in a new tab)
Tirth Kanani avatar

code-review-graph

Local code graph that feeds AI coding assistants only the relevant files

  • AI coding agents
  • Developer tools
  • Agent skills and plugins
View on GitHub
Website

What it is

A Python tool that parses a repository with Tree-sitter into a SQLite graph of functions, classes, imports, calls and test links, then serves compact context to AI coding assistants over MCP. When code changes, it traces the blast radius of callers, dependents and tests so the assistant reads only affected files during reviews. Developers using Claude Code, Cursor, Codex and similar tools can set it up with a single command, after which hooks or a file watcher refresh the graph incrementally as code changes.

Images

Facts

Stars
32,285 GitHub stars.
32,285

Forks
3,098
Commits
1,270
Contributors
128
Language
Python
License
MIT
Last commit
Oct 15, 2026
On GitHub since
February 2026
First release
February 2026
Platforms
macOS, Windows, Linux, CLI
Self-hostable
No
Author
Tirth Kanani
Tags
  • static-analysis
  • mcp-server
  • code-review
  • context-engineering
  • knowledge-graph
  • code-intelligence

Last synced from GitHub on Oct 20, 2026

The 10 best overlooked open-source projects, every Monday.

Share and more

Comments

No comments yet. Start the conversation.

Similar projects

  • Vitali Avagyan avatar

    Code-Graph-RAG

    Maps a codebase into a knowledge graph to query and edit in plain English

    5.3k
  • Martin Vogel avatar

    codebase-memory-mcp

    Local MCP server that indexes code into a queryable knowledge graph

    47k
  • Colby Mchenry avatar

    CodeGraph

    Local code graph that gives AI coding agents structural context via MCP

    74.5k