HomeProjectsArticlesPK BoardAbout
Explore Projects

codegraph

Active
GitHub C MIT

Description

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.

Key Features

  • 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
  • Dramatically reduces token consumption by giving LLMs precise context
  • Multi-language support covering AST parsing and dependency graph construction for major languages

Use Cases

💡 Provide a RAG alternative for large codebases so LLMs no longer read everything
💡 Inject precise project structure context when launching Claude Code or Codex
💡 Speed up onboarding with second-level answers about modules and dependencies
💡 Expose the code graph as an MCP server so any agent can query it

Strengths & Limitations

✅ Strengths

  • • Actively maintained, recent updates
  • • High community interest (66.6k stars)
  • • Permissive open-source license (MIT)
  • • Responsive to issues, low backlog

Tags

code-intelligence mcp rag local-first developer-tools

Categories

💻 Coding Agent 📚 RAG Tools

Quick Start

# Install the CodeGraph indexer
npm install -g @codegraph/cli

# Initialize index in project root
codegraph init

# Start MCP server for agent queries
codegraph serve --mcp

# Use it in Claude Code
> Find all callers of the user module
Visit GitHub

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

code-review-graph

30.4k · Python
Active A

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.

code-reviewknowledge-graphmcp +3
  • · 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
View full details → View on GitHub

UltraRAG

5.7k · Python
Active A+

A low-code MCP framework for building complex and innovative RAG pipelines. Combines visual pipeline design with MCP protocol integration for end-to-end RAG — from data ingestion and chunking to retrieval and generation.

ragmcppython +3
  • · First lightweight RAG framework built on MCP architecture
  • · Low-code YAML configuration for complex RAG workflow orchestration
  • · Core components (retriever, generator, etc.) as independent MCP Servers
View full details → View on GitHub