vitali87/code-graph-rag

Health v2 · maintenance only

Adoption not part of the score

4828 stars · 638 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

Code-Graph-RAG is a Python CLI tool and MCP server that parses multi-language codebases with Tree-sitter, builds a knowledge graph of code structure in Memgraph, and enables natural-language querying, editing, and optimization of code. It supports 14 languages under one unified graph schema and integrates with AI assistants like Claude Code.

Use cases

• understand a large multi-language monorepo with AI
• ask natural language questions about codebase structure
• find which functions handle authentication across services
• trace data flow and taint through code
• detect dead code and duplicate code
• structurally search and replace code by AST pattern
• integrate codebase intelligence into Claude Code via MCP

When to choose

• you maintain a large monorepo spanning multiple languages
• grep and IDE find-references can't answer cross-language structural questions
• you want an MCP server giving AI assistants deep codebase context
• you need call-chain, dependency, or data-flow analysis as graph queries

When to avoid

• you have a small single-language project where grep and IDE tooling suffice
• you don't want to run Memgraph and Qdrant infrastructure locally
• your language is unsupported (e.g. Scala is still in development)
• you need fully offline analysis without any LLM provider

Facets

cli-tool · maturity active

rag parser search-engine nlp llm-inference mcp developer-tools code-review developer-tools large-language-models parsers programming-languages python cli cross-platform knowledge-graph tree-sitter memgraph code-analysis monorepo ast cypher semantic-search mcp-server static-analysis retrieval-augmented-generation docker

Member repositories

For agents

markdown · JSON · MCP: product_card(name="vitali87/code-graph-rag")