HomeProjectsArticlesPK BoardAbout
Explore Projects

code-review-graph

Active
GitHub Python MIT

Description

Local-first code intelligence graph for MCP and CLI. Builds a persistent map of the codebase so AI coding tools read only what matters, with benchmarked context reductions on code review and large-repo workflows.

Key Features

  • Local-first — all indexing happens on the developer's machine, no data leaves the host
  • Code intelligence graph — parses code with tree-sitter and builds an entity-relation graph
  • MCP integration — exposes Model Context Protocol endpoints for Claude Code, Cursor, and other agents
  • Context compression — significantly reduces context consumption on code review and large-repo tasks
  • Incremental updates — listens for file changes and updates graph nodes on demand
  • Static analysis — graph-based dependency analysis supports impact-domain evaluation

Use Cases

💡 Giving Claude Code, Cursor, Codex, and other AI coding agents a map of the codebase
💡 Feeding only the relevant code slices to LLMs during code review on large repos
💡 Letting any MCP agent query file-level dependencies and impact ranges
💡 Providing graph-based smart navigation and code insights in the IDE
💡 Monitoring how code changes affect the system architecture

Strengths & Limitations

✅ Strengths

  • • Actively maintained, recent updates
  • • High community interest (30.4k stars)
  • • Permissive open-source license (MIT)
  • • Responsive to issues, low backlog

Tags

code-review knowledge-graph mcp static-analysis tree-sitter ai-coding

Categories

💻 Coding Agent 📚 RAG Tools

Quick Start

pip install code-review-graph
crg init /path/to/repo
crg serve --mcp  # start the MCP server
# In Claude Code / Cursor, register crg as an MCP server
Visit GitHub Visit Website

Related Projects

Semble

5.9k · Python
Active A+

Fast and accurate code search engine designed for AI agents. Uses semantic indexing instead of grep+read, reducing token consumption by roughly 98% while delivering higher-quality retrieval for code navigation tasks.

code-searchembeddingsmcp +2
  • · Lightning-fast semantic search — embedding-based code retrieval that reduces token usage by ~98% vs grep+read
  • · MCP server — exposes Model Context Protocol endpoints so agents can call it directly
  • · High recall — outperforms traditional text matching on multiple code-retrieval benchmarks
View full details → View on GitHub

codegraph

66.6k · C
Active A

CodeGraph is a context graph for coding agents, mapping how a codebase is wired together so LLM-driven tools can navigate dependencies and produce more accurate edits.

code-intelligencemcprag +2
  • · Pre-indexed code knowledge graph that auto-syncs code changes with millisecond queries
  • · Compatible with multiple coding agents including Claude Code, Codex, Cursor, and AntiGravity
  • · Runs 100% locally with indexes and queries kept on your machine
View full details → View on GitHub

MonkeyCode

4.4k · TypeScript
Active A

An AI coding platform for team collaboration, open-sourced by Chaitin, supporting code generation, review, and automated testing integration.

ai-codingteam-collaborationcode-review +2
  • · Code generation - generate complete code blocks and functions based on natural language description
  • · Code review - auto-review PRs for potential issues, security vulnerabilities, performance bottlenecks
  • · Automated testing - generate unit tests, integration test cases
View on GitHub

Related Articles

Code ReviewLLMPR-Agent

Code Review Agents: Static Analysis Plus LLM Review in Practice

A code review Agent is not just "let the LLM see the diff." This article systematically explains the layered architecture (static analysis, pattern matching, LLM review, context-aware), PR-Agent self-hosting, context design, false-positive suppression, and team collaboration models -- with precision metrics and adoption paths.

Read more →