code-graph-rag and flexible-graphrag

These are complementary tools that operate at different scales: code-graph-rag specializes in monorepo code understanding with knowledge graphs, while flexible-graphrag provides a broader infrastructure for building graph-based RAG systems across multiple database and data source options, making it suitable as a foundation that could integrate code analysis tools like the former.

code-graph-rag
67
Established
flexible-graphrag
65
Established
Maintenance 25/25
Adoption 10/25
Maturity 9/25
Community 23/25
Maintenance 13/25
Adoption 14/25
Maturity 18/25
Community 20/25
Stars: 2,072
Forks: 349
Downloads:
Commits (30d): 402
Language: Python
License: MIT
Stars: 110
Forks: 26
Downloads: 206
Commits (30d): 0
Language: Python
License: Apache-2.0
No Package No Dependents
No risk flags

About code-graph-rag

vitali87/code-graph-rag

The ultimate RAG for your monorepo. Query, understand, and edit multi-language codebases with the power of AI and knowledge graphs

Uses Tree-sitter for multi-language AST parsing and Memgraph for graph-based storage of codebase structure and relationships. Supports cloud models (Gemini, OpenAI), local models (Ollama), and language-agnostic Cypher queries to retrieve code snippets and enable surgical file editing with AST-based function targeting.

About flexible-graphrag

stevereiner/flexible-graphrag

Flexible GraphRAG: Python, LlamaIndex, Docker Compose: 8 Graph dbs, 10 Vector dbs, OpenSearch, Elasticsearch, Alfresco. 13 data sources (9 auto-sync), KG auto-building, schemas, LLMs, Docling or LlamaParse doc processing, GraphRAG, RAG only, Hybrid search, AI chat. React, Vue, Angular frontends, FastAPI backend, REST API, MCP Server. Please 🌟 Star

Supports RDF-based ontologies and SPARQL queries across both property graph and triple store databases, enabling schema-guided knowledge graph extraction. Implements automatic incremental synchronization that detects changes across 13 data sources and updates vector, search, and graph databases in near real-time without full reprocessing. Built on LlamaIndex abstractions with OpenTelemetry instrumentation for production observability, plus an MCP Server integration for Claude Desktop and compatible clients.

Scores updated daily from GitHub, PyPI, and npm data. How scores work